Redundent virtual link aggregation group

Information

  • Patent Grant
  • 10476698
  • Patent Number
    10,476,698
  • Date Filed
    Wednesday, March 18, 2015
    9 years ago
  • Date Issued
    Tuesday, November 12, 2019
    4 years ago
Abstract
One embodiment of the present invention provides a switch. The switch includes a link aggregation module and a link management module. The link aggregation module establishes a virtual link aggregation group comprising a plurality of links coupled to the switch and one or more other switches. The plurality of links includes a first and a second sets of links coupling a first and a second end devices, respectively. The link management module determines a current mode which indicates which of the sets of links is currently active, and operates the first and the second sets of links as active and standby links, respectively, for the virtual link aggregation group based on the current mode and a port role for a port participating in the virtual link aggregation group. The port role indicates whether the port is coupled to an active link or a backup link.
Description
BACKGROUND

Field


The present disclosure relates to network management. More specifically, the present disclosure relates to a method and system for facilitating device-level redundancy in a link aggregation group.


Related Art


The relentless growth of the Internet has brought with it an insatiable demand for bandwidth. As a result, equipment vendors race to build larger, faster, and more versatile switches to move traffic. However, the size of a switch cannot grow infinitely. It is limited by physical space, power consumption, and design complexity, to name a few factors. More importantly, because an overly large system often does not provide economy of scale due to its complexity, simply increasing the size and throughput of a switch may prove economically unviable due to the increased per-port cost.


A flexible way to improve the scalability of a switch system is to build a fabric switch. A fabric switch is a collection of individual member switches. These member switches form a single, logical switch that can have an arbitrary number of ports and an arbitrary topology. As demands grow, customers can adopt a “pay as you grow” approach to scale up the capacity of the fabric switch.


Meanwhile, layer-2 and layer-3 (e.g., Ethernet and Internet Protocol (IP), respectively) switching technologies continue to evolve. IP facilitates routing and end-to-end data transfer in wide area networks (WANs) while providing safeguards for error-free communication. On the other hand, more routing-like functionalities are migrating into layer-2. Notably, the recent development of the Transparent Interconnection of Lots of Links (TRILL) protocol allows Ethernet switches to function more like routing devices. TRILL overcomes the inherent inefficiency of the conventional spanning tree protocol, which forces layer-2 switches to be coupled in a logical spanning-tree topology to avoid looping. TRILL allows routing bridges (RBridges) to be coupled in an arbitrary topology without the risk of looping by implementing routing functions in switches and including a hop count in the TRILL header.


As more mission-critical applications are being implemented in data communication networks, high-availability operation is becoming progressively more important as a value proposition for network architects. It can be desirable to divide a conventional aggregated link (from one device to another) among multiple network devices, often belonging to different fabric switches, such that unavailability of one fabric switch would not affect the operation of the multi-homed device.


While a link aggregation brings many desirable features to a network, some issues remain unsolved in facilitating device-level redundancy in a virtual link aggregation group. Particularly, when a plurality of member switches of a fabric switch couple both active and standby end devices via a virtual link aggregation group, existing technologies do not provide a scalable and flexible solution that takes full advantage of the virtual link aggregation group.


SUMMARY

One embodiment of the present invention provides a switch. The switch includes a link aggregation module and a link management module. The link aggregation module establishes a virtual link aggregation group comprising a plurality of links coupled to the switch and one or more other switches. The plurality of links includes a first set of links coupling a first end device and a second set of links coupling a second end device. The link management module determines a current mode, which indicates which of the sets of links is currently active, of the virtual link aggregation group. The link management module operates the first set of links as active links carrying traffic for the virtual link aggregation group and the second set of links as standby links for the first set of links based on the current mode and a port role of a port participating in the virtual link aggregation group. The port role indicates whether the port is coupled to an active link or a backup link.


In a variation on this embodiment, the link aggregation module identifies an acknowledgment of a notification message from a remote switch of the other switches. The notification message includes port information associated with a local port participating in the virtual link aggregation group. Upon receiving the acknowledgment from a respective of the other switches, the link management module selects the first or second set of links for actively carrying traffic.


In a variation on this embodiment, the link management module determines the current mode by comparing a respective number of operational links in the first and second set of links with a protection threshold value.


In a further variation, the comparison comprises determining whether the number of operational links in one of the sets of links is lower than the protection threshold value and whether the number of operational links in another of the sets of links is greater than or equal to the protection threshold value.


In a variation on this embodiment, the link aggregation module maintains a database for the virtual link aggregation group. A respective entry in the database is associated with a port participating the virtual link aggregation group and includes a port role for the port.


In a further variation, if the port role indicates that the port is coupled to an active link and the current mode indicates that the first set of links is actively carrying traffic, the link management module marks the entry as selected to carry traffic.


In a further variation, if the port role indicates that the port is coupled to an active link and the current mode indicates that the second set of links is actively carrying traffic, the link management module marks the entry as standby.


In a variation on this embodiment, the current mode indicates that the second set of links is currently active. The link management module then operates the second set of links as active links carrying traffic for the virtual link aggregation group.


In a variation on this embodiment, the virtual link aggregation group is represented as a virtual switch identifier associated with the switch and the other switches.


In a further variation, the switch also includes a forwarding module which determines whether a local port participating in the virtual link aggregation group is coupled to a link carrying traffic for the virtual link aggregation group. If the port is coupled to a link carrying traffic, the forwarding module determines the port as an egress port of a packet, which is encapsulated in a header with the virtual switch identifier as the egress switch identifier.


In a further variation, if no local port is coupled to an operational link carrying traffic for the virtual link aggregation group, the forwarding module determines an inter-switch port as an egress port for the packet. The inter-switch port is associated with one of the other switches.


In a variation on this embodiment, the switch and the other switches are member switches of an Ethernet fabric switch, wherein the Ethernet fabric switch operates as a single Ethernet switch.





BRIEF DESCRIPTION OF THE FIGURES


FIG. 1A illustrates an exemplary redundant virtual link aggregation group, in accordance with an embodiment of the present invention.



FIG. 1B illustrates an exemplary redundant virtual link aggregation group with a virtual switch, in accordance with an embodiment of the present invention.



FIG. 2A illustrates an exemplary data structure in a switch for a redundant virtual link aggregation group, in accordance with an embodiment of the present invention.



FIG. 2B illustrates an exemplary state machine for a redundant virtual link aggregation group, in accordance with an embodiment of the present invention.



FIG. 3 illustrates an exemplary distributed initialization of a redundant virtual link aggregation group, in accordance with an embodiment of the present invention.



FIG. 4A presents a flowchart illustrating the process of a switch selecting an initial local port status for a redundant virtual link aggregation group, in accordance with an embodiment of the present invention.



FIG. 4B presents a flowchart illustrating the process of a switch selecting an initial remote port status for a redundant virtual link aggregation group, in accordance with an embodiment of the present invention.



FIG. 5A illustrates exemplary high availability in a redundant virtual link aggregation group, in accordance with an embodiment of the present invention.



FIG. 5B illustrates an exemplary data structure with selected active links in a redundant virtual link aggregation group in response to a failure, in accordance with an embodiment of the present invention.



FIG. 5C illustrates an exemplary data structure with selected standby links in a redundant virtual link aggregation group in response to a failure, in accordance with an embodiment of the present invention.



FIG. 5D illustrates an exemplary data structure with selected active links in a redundant virtual link aggregation group in response to a failure recovery, in accordance with an embodiment of the present invention.



FIG. 6A presents a flowchart illustrating the process of a switch selecting a local port status for a redundant virtual link aggregation group in response to a state change, in accordance with an embodiment of the present invention.



FIG. 6B presents a flowchart illustrating the process of a switch selecting a remote port status for a redundant virtual link aggregation group in response to a state change, in accordance with an embodiment of the present invention.



FIG. 7A presents a flowchart illustrating the process of a switch forwarding a packet received via an inter-switch port, in accordance with an embodiment of the present invention.



FIG. 7B presents a flowchart illustrating the process of a switch forwarding a packet received via an edge port participating in a redundant virtual link aggregation group, in accordance with an embodiment of the present invention.



FIG. 8 illustrates an exemplary participant switch of a redundant virtual link aggregation group, in accordance with an embodiment of the present invention.





In the figures, like reference numerals refer to the same figure elements.


DETAILED DESCRIPTION

The following description is presented to enable any person skilled in the art to make and use the invention, and is provided in the context of a particular application and its requirements. Various modifications to the disclosed embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other embodiments and applications without departing from the spirit and scope of the present invention. Thus, the present invention is not limited to the embodiments shown, but is to be accorded the widest scope consistent with the claims.


Overview

In embodiments of the present invention, the problem of facilitating device-level redundancy in a virtual link aggregation group (VLAG) is solved by dividing the links in the virtual link aggregation group into active and standby link sets for active and standby devices, respectively.


An end device (e.g., a host machine or a customer switch) can be coupled to a plurality of switches via a virtual link aggregation group. The plurality of switches participating in the virtual link aggregation group can be referred to as “participant switches” of the virtual link aggregation group. A port participating in the virtual link aggregation group can be referred to as a participant port. One or more of the participant switches can be a member switch of a fabric switch. With existing technologies, a respective participant switch operates a respective local participant port (and the link coupled to the port) as an active port for the virtual link aggregation group. This facilitates forwarding of different packets via different participant switches.


However, a user (e.g., a customer) can deploy active-standby redundancy among the end devices (e.g., customer switches) coupled via the virtual link aggregation group. As a result, the client may couple both active and standby end devices to the same virtual link aggregation group. If a link to the active end device fails, a participant switch may start forwarding data to a standby end device even when other links to the active end device remain operational. This can cause repeated network changes, which may lead to degraded performance. Furthermore, among the participant switches, one switch typically operates as the principal switch, which maintains the state of a respective participant switch and a participant port. During failover, a respective participant switch sends and receives control messages to and from this principal switch, respectively, to determine the current state of the virtual link aggregation group. Hence, this principal switch can become a single point of failure.


To solve this problem, a virtual link aggregation group includes a set of active links and a set of standby links. One set of links is allowed to carry traffic in the virtual link aggregation group at a time. The set of active links couples at least two participant switches to an active end device. Similarly, the set of standby links couples at least two participant switches to a standby end device. Such a link aggregation group can be referred to as a redundant link aggregation group (RVLAG). It should be noted that a redundant link aggregation group can couple more than one standby end devices. Under such circumstances, the set of standby links can include one or more subsets of standby links, and each subset of standby links couples a standby end device.


A respective participant switch of the redundant virtual link aggregation group maintains a data structure, which can be referred to as a redundant virtual link aggregation group database (RVLAG database). A respective entry of the database is associated with a link in the redundant virtual link aggregation group and indicates to which set the link belongs. The set of active links carries traffic for the redundant virtual link aggregation group as long as a minimum number of links in that set remain operational. During regular operation, a respective participant switch individually selects the set of active links to carry traffic (e.g., send or receive traffic) for the redundant virtual link aggregation group. If a minimum number of links in the set of active links is not operational, a respective participant switch individually makes a distributed decision to select the set (or a subset) of standby links to carry traffic based on the entries of the database. As a result, disruption to the network is reduced by providing high availability to the links of a link set, thereby reducing data loss during a failover. In this way, a redundant virtual link aggregation group provides device-level redundancy in the network.


In some embodiments, a participant switch can be in a fabric switch. A respective switch in a fabric switch can be referred to as a member switch. In a fabric switch, any number of switches coupled in an arbitrary topology may logically operate as a single switch. The fabric switch can be an Ethernet fabric switch or a virtual cluster switch (VCS), which can operate as a single Ethernet switch. Any member switch may join or leave the fabric switch in “plug-and-play” mode without any manual configuration. In some embodiments, a respective switch in the fabric switch is a Transparent Interconnection of Lots of Links (TRILL) routing bridge (RBridge). In some further embodiments, a respective switch in the fabric switch is an Internet Protocol (IP) routing-capable switch (e.g., an IP router).


In some embodiments, a respective member switch of the fabric switch can be equipped with a persistent storage framework, which stores the configuration information in a local persistent storage. Such a persistent storage can be an object relational database. The configuration information is loaded from this persistent storage to the switch (or device) modules (e.g., the application-specific integrated circuit (ASIC) chips of the switch). In some embodiments, an Object-Relational Mapping is used to store the attribute values of a switch unit in a structured way in an object relational database. When a unit becomes operational on the switch, attribute values associated with a respective class in that unit is automatically loaded from the database. Moreover, if a class changes (e.g., a new attribute or a new relationship), that change is seamlessly incorporated into the database.


It should be noted that a fabric switch is not the same as conventional switch stacking. In switch stacking, multiple switches are interconnected at a common location (often within the same rack), based on a particular topology, and manually configured in a particular way. These stacked switches typically share a common address, e.g., an IP address, so they can be addressed as a single switch externally. Furthermore, switch stacking requires a significant amount of manual configuration of the ports and inter-switch links. The need for manual configuration prohibits switch stacking from being a viable option in building a large-scale switching system. The topology restriction imposed by switch stacking also limits the number of switches that can be stacked. This is because it is very difficult, if not impossible, to design a stack topology that allows the overall switch bandwidth to scale adequately with the number of switch units.


In contrast, a fabric switch can include an arbitrary number of switches with individual addresses, can be based on an arbitrary topology, and does not require extensive manual configuration. The switches can reside in the same location, or be distributed over different locations. These features overcome the inherent limitations of switch stacking and make it possible to build a large “switch farm,” which can be treated as a single, logical switch. Due to the automatic configuration capabilities of the fabric switch, an individual physical switch can dynamically join or leave the fabric switch without disrupting services to the rest of the network.


Furthermore, the automatic and dynamic configurability of the fabric switch allows a network operator to build its switching system in a distributed and “pay-as-you-grow” fashion without sacrificing scalability. The fabric switch's ability to respond to changing network conditions makes it an ideal solution in a virtual computing environment, where network loads often change with time.


In switch stacking, multiple switches are interconnected at a common location (often within the same rack), based on a particular topology, and manually configured in a particular way. These stacked switches typically share a common address, e.g., an IP address, so they can be addressed as a single switch externally. Furthermore, switch stacking requires a significant amount of manual configuration of the ports and inter-switch links. The need for manual configuration prohibits switch stacking from being a viable option in building a large-scale switching system. The topology restriction imposed by switch stacking also limits the number of switches that can be stacked. This is because it is very difficult, if not impossible, to design a stack topology that allows the overall switch bandwidth to scale adequately with the number of switch units.


It should also be noted that a fabric switch is distinct from a virtual local area network (VLAN). A fabric switch can accommodate a plurality of VLANs. A VLAN is typically identified by a VLAN tag. In contrast, the fabric switch is identified a fabric identifier (e.g., a VCS identifier), which is assigned to the fabric switch. A respective member switch of the fabric switch is associated with the fabric identifier. Furthermore, when a member switch of a fabric switch learns a media access control (MAC) address of an end device (e.g., via layer-2 MAC address learning), the member switch generates a notification message, includes the learned MAC address in the payload of the notification message, and sends the notification message to all other member switches of the fabric switch. In this way, a learned MAC address is shared among a respective member switch of the fabric switch.


The term “fabric switch” refers to a number of interconnected physical switches which form a single, scalable switch. These physical switches are referred to as member switches of the fabric switch. In a fabric switch, any number of switches can be connected in an arbitrary topology, and the entire group of switches functions together as one single, logical switch. This feature makes it possible to use many smaller, inexpensive switches to construct a large fabric switch, which can be viewed as a single logical switch externally. Although the present disclosure is presented using examples based on a fabric switch, embodiments of the present invention are not limited to a fabric switch. Embodiments of the present invention are relevant to any computing device that includes a plurality of devices operating as a single device.


The term “end device” can refer to any device external to a fabric switch. Examples of an end device include, but are not limited to, a host machine, a conventional layer-2 switch, a layer-3 router, or any other type of network device. Additionally, an end device can be coupled to other switches or hosts further away from a layer-2 or layer-3 network. An end device can also be an aggregation point for a number of network devices to enter the fabric switch. An end device hosting one or more virtual machines can be referred to as a host machine. In this disclosure, the terms “end device” and “host machine” are used interchangeably.


The term “switch” is used in a generic sense, and it can refer to any standalone or fabric switch operating in any network layer. “Switch” should not be interpreted as limiting embodiments of the present invention to layer-2 networks. Any device that can forward traffic to an external device or another switch can be referred to as a “switch.” Any physical or virtual device (e.g., a virtual machine/switch operating on a computing device) that can forward traffic to an end device can be referred to as a “switch.” Examples of a “switch” include, but are not limited to, a layer-2 switch, a layer-3 router, a TRILL RBridge, or a fabric switch comprising a plurality of similar or heterogeneous smaller physical and/or virtual switches.


The term “edge port” refers to a port on a fabric switch which exchanges data frames with a network device outside of the fabric switch (i.e., an edge port is not used for exchanging data frames with another member switch of a fabric switch). The term “inter-switch port” refers to a port which sends/receives data frames among member switches of a fabric switch. The terms “interface” and “port” are used interchangeably.


The term “switch identifier” refers to a group of bits that can be used to identify a switch. Examples of a switch identifier include, but are not limited to, a media access control (MAC) address, an Internet Protocol (IP) address, and an RBridge identifier. Note that the TRILL standard uses “RBridge ID” (RBridge identifier) to denote a 48-bit intermediate-system-to-intermediate-system (IS-IS) System ID assigned to an RBridge, and “RBridge nickname” to denote a 16-bit value that serves as an abbreviation for the “RBridge ID.” In this disclosure, “switch identifier” is used as a generic term, is not limited to any bit format, and can refer to any format that can identify a switch. The term “RBridge identifier” is also used in a generic sense, is not limited to any bit format, and can refer to “RBridge ID,” “RBridge nickname,” or any other format that can identify an RBridge.


The term “packet” refers to a group of bits that can be transported together across a network. “Packet” should not be interpreted as limiting embodiments of the present invention to layer-3 networks. “Packet” can be replaced by other terminologies referring to a group of bits, such as “message,” “frame,” “cell,” or “datagram.”


The term “dual-homed end device” refers to an end device that has an aggregate link to two or more switches belonging to one or more fabric switches, where the aggregate link includes multiple physical links to the different switches. The aggregate link, which includes multiple physical links, functions as one logical link to the end station. Although the term “dual” is used here, the term “dual-homed end device” does not limit the number of physical switches sharing the aggregate link to two. In various embodiments, other numbers of physical switches can share the same aggregate link. Where “dual-homed end device” is used in the present disclosure, the term “multi-homed end device” can also be used.


Network Architecture



FIG. 1A illustrates an exemplary redundant virtual link aggregation group, in accordance with an embodiment of the present invention. As illustrated in FIG. 1A, a fabric switch 100 includes switches 101, 102, 103, 104, and 105. A switch in a fabric switch can be referred to as a member switch. A member switch, such as switch 102, of fabric switch 100 can be configured by logging in to switch 102 (e.g., via telnet) or via a console port (e.g., an RS-232 port). Such configuration can be related to network virtualizations, partitions, and switch groups, and a plurality of network protocols of different network layers. The attribute values (e.g., parameters) of the configuration information can be stored in a local persistent storage and applied to switch 102 (e.g., loaded to the switch modules). Configuration and state information of fabric switch 100 can be stored in a persistent storage of a respective member switch of fabric switch group 100.


An end device coupled to fabric switch 100 can be a host machine (e.g., a server or a computing device hosting virtual machines) or a customer networking device (e.g., a layer-2 switch or a layer-3 routing device). In this example, an end device 112, which is a host machine, is coupled to switch 102. End device 112 can host one or more virtual machines. End device 112 can include a hypervisor, which runs one or more virtual machines. End device 112 can be equipped with a Network Interface Card (NIC) with one or more ports. End device 112 couples to switch 102 via the ports of the NIC. On the other hand, end devices 122 and 124, which are coupled to fabric switch 100, are customer network devices. End devices 122 and 124 can be coupled to end device 114, which can be a host machine. Since end devices 122 and 124 are customer network devices, they can forward traffic received from fabric switch 100 to end device 114. In this disclosure, the terms “end device” and “customer network device” are used interchangeably.


In some embodiments, data communication among the member switches of fabric switch 100 is based on the TRILL protocol and a respective switch of fabric switch 100, such as switch 102, is a TRILL RBridge. Upon receiving an Ethernet frame from end device 112, switch 102 encapsulates the received Ethernet frame in a TRILL header and forwards the TRILL packet. In some embodiments, data communication among the member switches of fabric switch 100 is based on IP and a respective switch of fabric switch 100, such as switch 102, is an IP-capable switch. An IP-capable switch populates and maintains a local IP routing table, such as a routing information base, or RIB, by participating in a routing algorithm, and is capable of forwarding packets based on its IP addresses. For example, upon receiving an Ethernet frame from end device 112, switch 102 encapsulates the received Ethernet frame in an IP header and forwards the IP packet.


Switches 103, 104, and 105 are configured to operate in a special “trunked” mode for dual-homed end devices 122 and 124 and form a virtual link aggregation group 130. Switches 103, 104, and 105 can be referred to as “participant switches” of virtual link aggregation group 130. Switch identifiers 151, 152, and 153 are associated with switches 103, 104, and 105, respectively. Switch identifiers 151, 152, and 153 uniquely identify switches 103, 104, and 105, respectively, in fabric switch 100. In some embodiments, the scope of switch identifiers 151, 152, and 153 is within fabric switch 100. Ports 131 and 132 of switch 103 participate in virtual link aggregation group 130 and can be referred to as participant ports. Similarly, ports 133 and 134 of switch 104 and ports 135 and 136 of switch 105 are participant ports of virtual link aggregation group 130 as well. Port identifiers 161, 162, 163, 164, 165, and 166 are associated with participant ports 131, 132, 133, 134, 135, and 136, respectively.


With existing technologies, participant switch 105 operates ports 135 and 136 as active ports for virtual link aggregation group 130. Similarly, switches 103 and 104 operate their respective local participant ports as active ports as well. This facilitates forwarding of different packets via different participant switches. However, a user (e.g., a customer) can deploy active-standby redundancy among end devices 122 and 124 coupled via virtual link aggregation group 130. Suppose that end device 122 is the active device and end device 124 is the standby device. If a link to end device 122 (e.g., the link coupled to port 135) fails, switch 105 may start forwarding data to end device 124 even when other links to end device 122 (e.g., links coupled to ports 131-133) remain operational. This can cause repeated network changes, which may lead to degraded performance.


Furthermore, among participant switches 103, 104, and 105, one switch typically operates as the principal switch and maintains the state of a respective participant switch and a respective participant port. During a failover, a respective participant switch sends and receives control messages to and from this principal switch, respectively, to determine the current state of virtual link aggregation group 130. This principal switch can become a single point of failure and create a bottleneck.


To solve this problem, virtual link aggregation group 130 includes a set 192 of active links and a set 194 of standby links. One of these sets of links is allowed to carry traffic in virtual link aggregation group 130 at a time. Set 192 couple participant switches 103, 104, and 105 to end device 122. Similarly, set 194 couple participant switches 104 and 105 to end device 124. Such a link aggregation group can be referred to as a redundant link aggregation group. It should be noted that redundant link aggregation group 130 can couple more than one standby end devices. Under such circumstances, set 194 can include one or more subsets of standby links, and each subset of standby links couple a standby end device. In some embodiments, virtual link aggregation group 130 is associated with a group identifier, which is the same in a respective participant switch. A respective participant switch identifies virtual link aggregation group 130 by that group identifier and can maintain a mapping between the local participant ports and the group identifier.


Switches 103, 104, and 105 maintain a data structure 142, 144, and 146, respectively, to store information associated with redundant virtual link aggregation group 130. This data structure can be referred to as a redundant virtual link aggregation group database (RVLAG database). For example, a respective entry of database 142 corresponds to a port coupled to a link in redundant virtual link aggregation group 130 and is represented by the port identifier of the port. The entry can also include the switch identifier of the switch which includes the port. This entry further indicates whether the link belongs to set 192 or set 194. For example, the entry for the link between switch 104 and end device 122 can include port identifier 165 and switch identifier 153, and indicates that the link belong to set 192.


During initialization, switch 103, 104, and 105 individually select set 192 to carry traffic for redundant virtual link aggregation group 130. Links in set 192 carry traffic as long as a minimum number of links in set 192 remain operational. For example, if the minimum number is two, links in set 192 carry traffic as long as at least two links in set 192 remain operational. When one of the participant switches, such as switch 105, receives a packet for end device 114, switch 105 determines that the packet should be forwarded via redundant virtual link aggregation group 130. Switch 105 forwards the packet via port 135 coupled to an active link. If the link becomes unavailable (e.g., due a failure), switch 105 does not have a local participant port coupled to an active link. Switch 105 then forwards the packet to another participant switch, such as switch 104, coupled to an operational active link. This allows switch 105 to forward the packet to end device 122 even when the link coupled to port 135 fails. As a result, disruption to the network is reduced by providing high availability to the links of set 192 (and set 194), thereby reducing data loss during a failover.


If the number of operational links in set 192 falls below two, switches 103, 104, and 105 individually select the links in set 194 to forward traffic based on the entries of in databases 142, 144, and 146, respectively. For example, switch 103 checks the entries of database 142 to determine that the number of operational links in set 192 has fallen below two and selects set 194 to forward traffic. Since the ports coupled to the links in set 194 are mapped to the same group identifier of redundant virtual link aggregation group 130, a respective participant switch can readily select the ports coupled to the links in set 194 for forwarding traffic. It should be noted that even though switch 103 can be coupled to an operational link of set 192 and not coupled to a link in set 194, switch 103 selects set 194 to forward traffic. Switches 104 and 105 individually select set 194 to forward traffic as well. In some embodiments, selection of set 194 is atomic among the participant switches of virtual link aggregation group 130. This atomic operation ensures that the operation is executed at a participant switch when all participant switches “agree” with (e.g., can also execute) the operation.


When switch 105 receives another packet which should be forwarded via redundant virtual link aggregation group 130, switch 105 forwards the packet via port 136, which is coupled to a link in set 194. When the links of set 192 recover from the failure and at least two links in set 192 become operational, switches 103, 104, and 105 can continue to use the links in set 194 to carry traffic. In this way, redundant virtual link aggregation group 130 can reduce the number of changes in the network. However, if the number of operational links in set 194 falls below two, switches 103, 104, and 105 individually select set 192 to forward traffic. In this way, redundant virtual link aggregation group 130 facilitates device-level high availability, which is switch-level high availability in this example, to end devices 122 and 124.



FIG. 1B illustrates an exemplary redundant virtual link aggregation group with a virtual switch, in accordance with an embodiment of the present invention. In this example, redundant virtual link aggregation group 130 is represented as a virtual switch 110. It should be noted that virtual switch 110 is distinct from a virtual customer network device, which can be associated with end devices 122 and 124. End devices 122 and 124 view participant switches 103, 104, and 105 as a common virtual switch 110, with a corresponding virtual switch identifier. Dual-homed end devices 122 and 124 are considered to be logically coupled to virtual switch 110 via logical links represented by dotted lines. Virtual switch 110 is considered to be logically coupled to participant switches 103, 104, and 105, optionally with zero-cost links (also represented by dotted lines). Furthermore, switches 103, 104, and 105 can advertise their respective connectivity to virtual switch 110. Hence, multi-pathing can be achieved when switches 101 and 102 choose to send packets to virtual switch 110 (which are marked as the egress switch in the packets) via switches 103, 104, and 105.


During operation, switch 102 learns the MAC address of end device 112 and distributes the learned MAC address in a payload of a notification message to a respective switch of fabric switch 100. Based on the notification, a respective switch of fabric switch maintains a mapping between the MAC address of end device 112 and the switch identifier of switch 102. When end device 114 sends a packet to end device 112 via active end device 122, one of the participant switches, such as switch 104, receives the packet. Switch 104 determines from its local mapping that end device 112 is reachable via switch 102. Switch 104 encapsulates the packet in a fabric encapsulation header (e.g., a TRILL or an IP header) and assigns the virtual switch identifier of virtual switch 110 as the ingress switch identifier of the encapsulation header. Switch 104 then forwards the packet to switch 102 via an inter-switch link. Upon receiving the packet, switch 102 learns that end device 114 is reachable via virtual switch 110.


Since participant switches 103, 104, and 105 function as a single virtual switch 110, the MAC address reachability learned by a participant switch is shared with the other participant switches of redundant virtual link aggregation group 130. For example, during normal operation, end device 122 may choose to send outgoing packets from end device 114 only via the links to switch 104. As a result, only switch 104 learns the MAC address of end device 114 and associates the MAC address with the virtual switch identifier of virtual switch 110. This information is then shared by switch 104 with switches 103 and 105. When end device 112 sends a packet to end device 114, switch 102 encapsulates the packet in an encapsulation header and assigns the virtual switch identifier of virtual switch 110 as the egress switch identifier of the encapsulation header.


Since virtual switch 110 is “reachable” via any of switches 103, 104, and 105, switch 102 can forward the packet to switch 103, 104, or 105. If switch 103 receives the packet, switch 103 determines that the egress switch identifier of the encapsulation header is associated with the local switch and decapsulates the fabric encapsulation header. Switch 103 extracts the inner packet, which can be an Ethernet frame with the MAC address of end device 114 as the destination MAC address. Switch 103 determines that the MAC address of end device 114 is associated with the virtual switch identifier of virtual switch 110, which is associated with redundant virtual link aggregation group 130. Switch 103 then forwards the packet via one of its participant ports 131 and 132. In some embodiments, switch 103 deploys a selection technique (e.g., round robin or weighted selection) to select one of the participant ports 131 and 132 for forwarding the packet.


Database and State Machine


In the example in FIG. 1A, participant switches 103, 104, and 105 maintain local databases 142, 144, and 146, respectively, comprising information of a respective participant port of redundant virtual link aggregation group 130. FIG. 2A illustrates an exemplary data structure in a switch for a redundant virtual link aggregation group, in accordance with an embodiment of the present invention. The data structure in switch 103 is redundant virtual link aggregation group database 142. Switch 103 can be coupled to an end device 116. A user (e.g., a network administrator) can provide configuration information associated with redundant virtual link aggregation group 130 from end device 116 by logging in to switch 103 (e.g., via telnet) or via a console port of switch 103 (e.g., an RS-232 port). In some embodiments, switch 103 includes an object relational database 220, which stores global configuration information associated with fabric switch 100 and local configuration information associated with switch 103. Database 142 can be stored as a table in object relational database 220.


A respective entry of database 142 corresponds to a link in redundant virtual link aggregation group 130. The entry can include one or more fields, such as a port identifier 201, a switch identifier 202, a port role 203, a port flag 204, and a port status 205. A respective field can be represented as a column of database 142. Port identifier 201 identifies a port which couples the link and can be used as the index for database 142. Switch identifier 202 identifies the switch in which the port resides. Port role 203 indicates whether a port belongs to set 192 or set 194 (e.g., active or backup) and can be preconfigured. Port flag 204 indicates whether information associated with a local participant port has been updated at other participant switches. Port status 205 indicates which set is currently selected to carry traffic for redundant virtual link aggregation group 130.


During operation, switch 103, which has switch identifier 151, discovers that its local port 131, which has port identifier 161, is a participant port of redundant virtual link aggregation group 130. Switch 103 adds the information in an entry comprising port identifier 161 in database 142. Switch 103 then sends a notification message comprising the port information to switches 104 and 105. Similarly, upon discovering that port 132, which has port identifier 162, is a participant port of redundant virtual link aggregation group 130, switch 103 adds the information in an entry comprising port identifier 162 in database 142. Switch 103 sends another notification message comprising the port information to switches 104 and 105. While waiting for an acknowledgement from switches 104 and 105, switch 103 sets the corresponding port flag to WAIT_ACK, which indicates that a notification for that port has been sent but an acknowledgement from a respective other participant switch has not been received yet.


In this example, upon receiving a respective acknowledgement for the notification for port identifier 161, switch 103 sets the corresponding port flag to ALL_ACK, which indicates that an acknowledgement from a respective other participant switch has been received. Upon setting the port flag, switch 103 determines the port status for port identifier 161 (i.e., port 131) based on its port role. Since port role associated with port identifier 161 is “active,” switch 103 sets the port status to be “selected,” which indicates that the port should carry traffic for redundant virtual link aggregation group 130. On the other hand, if the port flag remains in WAIT_ACK, switch 103 does not set a port status for port identifier 162 (i.e., for port 132). If switch 103 receives a respective acknowledgement for the notification for port identifier 162, switch 103 sets the corresponding port flag to ALL_ACK.


When switch 103 receives a notification message from switch 104, which has switch identifier 152, switch 103 discovers that port 133, which has port identifier 163, is a participant port of redundant virtual link aggregation group 130. Switch 103 adds the information in an entry comprising port identifier 161 in database 142. Since port 133 is in a remote participant switch 104, port 133 is a remote participant port for switch 103. Switch 103 sets the port flag to X, which indicates that a value of the port flag is not relevant to an entry for a remote participant port. Switch 103 then sends an acknowledgement message in response to the notification message to switch 104. Switch 103 then determines the port status for port identifier 163 (i.e., port 133) based on its port role. Since port role associated with port identifier 163 is “active,” switch 103 sets the port status to be “selected.”


Similarly, upon receiving a notification from switch 104, switch 103 adds the information associated with port 134, which has port identifier 164, in an entry comprising port identifier 164 in database 142. Switch 103 sets the port flag to X, sends an acknowledgement message to switch 104, and determines the port status for port identifier 164 (i.e., port 134). Since port role associated with port identifier 164 is “backup,” switch 103 sets the port status to be “standby.” In the same way, upon receiving respective notifications from switch 105, switch 103 adds the information associated with ports 135 and 136, which have port identifiers 165 and 166, respectively, in corresponding entries in database 142. Switch 103 sets the corresponding port flags to X, sends acknowledgement messages to switch 105 for respective notification messages, and determines the port status for port identifiers 165 and 166 (i.e., ports 135 and 136, respectively). Since port role associated with port identifiers 165 and 166 are “active” and “backup,” respectively, switch 103 sets the port status to be “selected” and “standby,” respectively.


In the example in FIG. 2A, when a minimum number of active links are selected (i.e., a minimum number of entries have a “selected” port status in database 142), switch 103 can initiate a state machine for redundant virtual link aggregation group 130. Similarly, switch 104 and 105 can individually initiate the state machine in a distributed way. FIG. 2B illustrates an exemplary state machine for a redundant virtual link aggregation group, in accordance with an embodiment of the present invention. Participant switches 103, 104, and 105 individually maintain a state machine 250 for redundant virtual link aggregation group 130. When a participant switch, such as switch 103, discovers that the local switch is in a redundant virtual link aggregation group 130, switch 103 initiates state machine 250 by starting in an init mode 260.


When switch 103 selects a minimum number of active links (i.e., marks a minimum number of entries as “selected” in database 142), switch 103 transitions from init mode 260 to an active mode 270. The minimum number of active links needed to transition from init mode 260 to active mode 270 is referred to as an initialization threshold. In some embodiments, the initialization threshold is one. In the same way, when switches 104 and 105 select a minimum number of active links from databases 144 and 146, respectively, switches 104 and 105 individually transition from init mode 260 to active mode 270. When a participant switch is in active mode 270, the active links (i.e., the links coupled to ports with an “active” port role in database 142) are selected to operate and carry traffic for redundant virtual link aggregation group 130. In the example in FIG. 1A, active links are in set 192. As long as a minimum number of links coupled to the ports with an “active” port role remain operational, a participant switch remains in active mode 270. The minimum number of links needed to remain in active mode 270 is referred to as a protection threshold.


If a link becomes unavailable (e.g., due to a failure), a “link down” link state change event occurs for that link. On the other hand, if an unavailable link becomes available (e.g., due to a failure recovery), a “link up” link state change event occurs for that link. A link state change event can be detected locally or received from a remote participant switch. When a participant switch, such as switch 103, detects a link state change event, switch 103 checks the number of active links and backup links (i.e., the number of links coupled to ports with a “backup” port role in database 142). If the number of active links falls below the protection threshold (e.g., due to a link down event) and the number of backup links is greater than or equal to the protection threshold, switch 103 transitions from active mode 270 to a protection mode 280. In the same way, switches 104 and 105 individually detect a link state change event, make the determinations, and transition from active mode 270 to protection mode 280.


In some embodiments, transitioning from active mode 270 to the protection mode 280 in a respective participant switch is atomic among the participant switches of redundant virtual link aggregation group 130. This atomic operation ensures that the operation is only executed at a participant switch when all participant switches “agree” with (e.g., can also execute) the operation. For example, when switch 103 determines to transition from active mode 270 to protection mode 280, switch 103 obtains a lock (e.g., a fabric-wide lock in fabric switch 100) for state machine 250. As a result, switches 104 and 105 do not transition between modes in state machine 250 as long as the lock is active. Switch 103 sends a control message to switches 104 and 105 indicating that switch 103 is ready for the transition.


Upon receiving the control message, switch 104 and 105 check whether this transition operation is allowed based on databases 144 and 146, respectively. If allowed, switches 104 and 105 send respective agreement messages back to switch 103. Upon receiving the agreement messages, switch 103 transitions from active mode 270 to protection mode 280 and releases the lock. If either switch 104 and 105 determines that the transition operation is not allowed, that switch sends a disagreement message to switch 103. If switch 103 receives a disagreement message, switch 103 cancels the transition operation and releases the lock. In this way, participant switches avoid a race condition during a transition in state machine 250.


When a participant switch, such as switch 103, is in the protection mode 280, the backup links are selected to operate and carry traffic for redundant virtual link aggregation group 130. In database 142, switch 103 sets the port status as “selected” to the entries with a “backup” port role. In the example in FIG. 1A, the backup links are in set 194. If switch 103 detects another link state change event, and determines that the number of active links has become greater than or equal to the protection threshold (e.g., due to a link up event) but the number of backup links remains greater than or equal to the protection threshold, switch 103 continues to use the backup links to carry traffic. This reduces changes in the network. However, if switch 103 determines that the number of active links is greater than or equal to the protection threshold and the number of backup links has fallen below the protection threshold, switch 103 transitions back from protection mode 280 to active mode 270. In the same way, switches 104 and 105 individually detect a link state change event, make the determinations, and transition back from active mode 270 to protection mode 280.


Initialization



FIG. 3 illustrates an exemplary distributed initialization of a redundant virtual link aggregation group, in accordance with an embodiment of the present invention. In this example, participant switches 103, 104, and 105 initializes redundant virtual link aggregation group 130. During operation, a respective participant switch (e.g., switch 103) detects redundant virtual link aggregation group 130 (operation 302). This detection can be based on a link aggregation detection protocol (e.g., Link Aggregation Control Protocol (LACP)) or from preconfigured information (e.g., configuration information provided by a user). The switch then identifies a respective local port participating in redundant virtual link aggregation group 130 (operation 304). The switch sends the local port information associated with a respective identified port to other participant switches and waits for an acknowledgement (operation 306). The port information includes one or more of: a port identifier, which can uniquely identify a port in a fabric switch, a switch identifier which identifies the switch in which the port resides, and a port role associated with the port.


Remote participant switches individually send their respective port information associated their local ports in redundant virtual link aggregation group 130 to the local participant switch as well. From the local switch's perspective, these ports are remote participant ports and their associated port information is remote participant port information. In the example in FIG. 1A, if the local switch is switch 103, switches 104 and 105 are the remote participant switches, and ports 131 and 132 are the local participant ports, and ports 133, 134, 135, and 136 are the remote participant ports. The local participant switch receives remote participant port information (e.g., via a notification message) from other participant switches (operation 308). The switch then adds the received remote participant port information to its local redundant virtual link aggregation group database and sends an acknowledgement for the received information to the other participant switches (operation 310).


The switch receives respective acknowledgements from other participant switches for its local port information and adds the local port information to its local database (operation 312). The switch then determines the mode for the redundant virtual link aggregation group based on the local database (operation 314). For example, the switch can determine the mode to be active based on whether the number of ports with an “active” port role is greater than or equal to an initialization threshold. Based on the determined mode, the switch marks the port status of a respective entry in the local database (operation 316). For example, if the mode is active, the switch marks the entries with an “active” port role as “selected” and with a “backup” port role as “standby.”



FIG. 4A presents a flowchart illustrating the process of a switch selecting an initial local port status for a redundant virtual link aggregation group, in accordance with an embodiment of the present invention. During operation, the switch identifies a local participant port in a redundant virtual link aggregation group (operation 402). The switch then checks whether the link coupled to the port is ready for initialization (operation 404). If the link is statically configured, the link is ready for initialization if the link is operational (e.g., can send and receive a ping via the link). If the link is dynamically configured (e.g., using LACP), the link is ready for initialization if the switch has received partner port information (i.e., the port coupled to the other side of the link). The link can also be ready for initialization if the link is in a defaulted mode, wherein the link can be ready without receiving partner port information.


If the link is not ready, the switch waits for the link to be ready (e.g., wait for partner port information) (operation 406) and continues to check whether the link is ready (operation 404). If the link is ready, the switch determines the port role of the identified port (operation 408). In some embodiments, a port role is predetermined. The port role can be predetermined based on a pre-configuration from a user. The port role can be also be determined based on a policy-based pre-computation, such as a predetermined number of ports with a superior (i.e., the highest or lowest) port identifier values are assigned one port role (e.g., “active”) and the rest of the ports are assigned another port role (e.g., “backup.”) The switch then adds the port information of the identified local port to the local redundant virtual link aggregation group database (operation 410). Adding port information to the local database includes creating an entry in the database and inserting values associated with different fields of the entry, as described in conjunction with FIG. 2A.


The switch constructs a notification message comprising the port information in its payload (operation 412). This notification message can be a fabric control message of a fabric switch. This fabric control message can be a reliable message with ensured delivery (e.g., retransmitted if the message is lost). The switch then sends the notification message to a respective other participant switch and sets a port flag of the local port to WAIT_ACK (operation 414). In some embodiments, sending a message includes identifying one or more local egress ports corresponding to the egress switch identifier of the message and transmitting the message via the identified port(s). The switch then receives an acknowledgement (e.g., an acknowledgement message) from a remote participant switch (operation 416). The switch checks whether an acknowledgement from a respective remote participant switch has been received (operation 418). If not, the switch continues to receive an acknowledgement from a remote participant switch (operation 416).


If the switch has received an acknowledgement from a respective remote participant switch, the switch sets the port flag to ALL_ACK (operation 420). The switch then runs the state machine for the redundant virtual link aggregation group based on the database, as described in conjunction with FIG. 2A, and determines the mode for the redundant virtual link aggregation group (operation 422). The switch marks the port status of a respective entry in the database based on the determined mode of the redundant virtual link aggregation group (operation 424). The switch identifies the local port(s) with a “selected” port status and aggregates the links coupled to the identified port(s) in the redundant virtual link aggregation group (operation 426). In some embodiments, the switch sends a confirmation message for a respective identified port to a respective remote participant switch (operation 428). This confirmation message indicates that the port is operational and ready to carry traffic for the redundant virtual link aggregation group. This confirmation message can be a fabric control message and can be sent for any link aggregation in a fabric switch.



FIG. 4B presents a flowchart illustrating the process of a switch selecting an initial remote port status for a redundant virtual link aggregation group, in accordance with an embodiment of the present invention. During operation, the switch receives a notification message comprising port information of a remote participant port of the redundant virtual link aggregation group from a remote participant switch (operation 452). The switch sends an acknowledgement message to the remote participant switch (operation 454) and adds this received port information to the local redundant virtual link aggregation group database (operation 456). The switch then runs the state machine for the redundant virtual link aggregation group based on the database, as described in conjunction with FIG. 2A, and determines the mode for the redundant virtual link aggregation group (operation 458).


The switch identifies port status of a respective entry in the database based on the determined mode of the redundant virtual link aggregation group (operation 460). In some embodiments, the switch receives a confirmation message from the remote participant switch indicating that the remote port is operational (operation 462). Upon receiving the confirmation message, the switch can mark this remote port to be ready to carry traffic for the redundant virtual link aggregation group (operation 464). This confirmation message can be a fabric control message and can be sent for any link aggregation in a fabric switch.


High Availability


In the example in FIG. 1A, redundant virtual link aggregation group 130 starts to carry traffic via the links in set 194 if a minimum number of links in set 192 are not operational, thereby facilitating high availability within the group. FIG. 5A illustrates exemplary high availability in a redundant virtual link aggregation group, in accordance with an embodiment of the present invention. In this example, the protection threshold is two for redundant virtual link aggregation group 130. Suppose that a failure 502 makes the links coupled to ports 132 and 133 unavailable. Upon detecting the unavailability of the link coupled to port 132, switch 103 notifies switches 104 and 105 regarding the unavailability. In the same way, upon detecting the unavailability of the link coupled to port 133, switch 104 notifies switches 103 and 105 regarding the unavailability.


Upon detecting a local failure or being notified regarding a remote failure, switches 103, 104, and 105 individually updates their respective redundant virtual link aggregation group database. At this point, the links coupled to ports 131 and 135 remain operational. As a result, the number of operational links in set 192 does not fall below the protection threshold and the participant switches continue to forward traffic via the operational links in set 192. Under such circumstances, if end device 112 sends a packet to end device 114, switch 102 receives the packet. Switch 102 encapsulates the packet in a fabric encapsulation header (e.g., a TRILL or an IP header) and assigns the virtual switch identifier of virtual switch 110 as the egress switch identifier of the encapsulation header and forwards it to virtual switch 110.


If switch 104 receives the packet, switch 104 determines that traffic is currently being carried by the operational links in set 192 and the local switch does not have an operational link in set 192. As a result, even though switch 104 is associated with virtual switch 110 (i.e., the virtual switch identifier of switch 110 is also associated with switch 104), switch 104 does not decapsulates the encapsulation header. Instead, switch 104 identifies that both switches 103 and 105 have local ports coupled to an operational link in set 192. Switch 104 then forwards the packet to either switch 103 or 105. Suppose that switch 103 receives the packet and determines that the local switch has port 131 coupled to an operational link in set 192. Hence, switch 103 decapsulates the encapsulation header and forwards the inner packet via port 131.


Suppose that another failure 504 makes the link coupled to port 135 unavailable. At that point, only the link coupled to port 131 remains operational in set 192. As a result, the number of operational links in set 192 has fallen below the protection threshold and set 192 does not have enough links to carry traffic to end device 122. Since the standby links (i.e., the links in set 194) are not affected by failures 502 and 504, the number of operational links in set 194 is greater than or equal to the protection threshold. Hence, the participant switches individually transition to the protection mode, as described in conjunction with FIG. 2B, and start forwarding traffic via the links in set 194. If switch 103 receives a packet destined to virtual switch 110, switch 103 determines that traffic is currently being carried by the links in set 194 and the local switch does not have a link in set 194. As a result, even though switch 103 is associated with virtual switch 110, instead of decapsulating, switch 103 identifies that both switches 104 and 105 have local ports coupled to a link in set 194. Switch 103 then forwards the packet to either switch 104 or 105. If switch 104 receives the packet, switch 104 decapsulates the encapsulation header and forwards the packet to end device 124.


At this point, since standby end device 124 has started receiving traffic, end device 124 can start operating as the active customer network device. End device 124 forwards this packet to end device 114. Upon receiving the packet, end device 114 relearns the MAC address of end device 112 via the port which couples end device 124. On the other hand, if end device 124 is associated with a virtual customer network device (e.g., based on a protection protocol, such as Virtual Router Redundancy Protocol (VRRP) or Virtual Switch Redundancy Protocol (VSRP)), upon becoming active, end device 114 relearns the MAC address of that virtual customer network device via the port which couples end device 124. In this way, when end device 114 forwards a packet to end device 112, end device 114 forwards the packet via the port which couples end device 124.


Suppose that a recovery from failure 502 makes the links coupled to ports 132 and 133 available. As a result, the number of operational links in set 192 becomes greater than or equal to the protection threshold. However, since the standby links of set 194 are currently selected and the number of operational links in set 194 is also greater than or equal to the protection threshold, the participant switches continue to forwarding traffic via the links in set 194. On the other hand, suppose that another failure 506 makes the link coupled to port 135 unavailable. At that point, the number of operational links in set 192 is greater than or equal to the protection threshold, but the number of operational links in set 194 has fallen below the protection threshold. As a result, the participant switches individually transition back to the active mode and start forwarding traffic via the operational links in set 192. These links are coupled to ports 131, 132, and 133.



FIG. 5B illustrates an exemplary data structure with selected active links in a redundant virtual link aggregation group in response to a failure, in accordance with an embodiment of the present invention. This example shows database 142 in response to failure 502, which makes ports 132 and 133 unavailable. Upon detecting a local failure 502 to port 132, switch 103 sends a notification to switches 104 and 105 regarding the unavailability of the link coupled to port 132, and waits for an acknowledgement from switches 104 and 105. Upon receiving a respective acknowledgement, switch 103 removes the entry comprising port identifier 162 of port 132 from database 142. Switch 103 runs state machine 250 based on database 142 and determines that redundant virtual link aggregation group 130 remains in active mode 270, as described in conjunction with FIG. 2B.


On the other hand, when switch 103 receives a notification from switch 104 regarding unavailability of port 133, switch 103 sends back an acknowledgement and removes the entry comprising port identifier 163 of port 133 from database 142. Switch 103 again runs state machine 250 based on database 142 and determines that redundant virtual link aggregation group 130 remains in active mode 270, as described in conjunction with FIG. 2B. As a result, the port status remains “selected” for the entries with an “active” port role and the port status remains “standby” for the entries with a “backup” port role in database 142. In the same way, switches 104, and 105 individually update their databases 144 and 146, respectively, and run state machine 250 to determine that that redundant virtual link aggregation group 130 remains in active mode 270.



FIG. 5C illustrates an exemplary data structure with selected standby links in a redundant virtual link aggregation group in response to a failure, in accordance with an embodiment of the present invention. This example shows database 142 in response to failures 502 and 504, which make ports 132, 133, and 135 unavailable. Upon detecting a local failure 504 to port 135, switch 105 sends a notification to switches 103 and 104 regarding the unavailability of the link coupled to port 135. When switch 103 receives a notification from switch 105 regarding unavailability of port 135, switch 103 sends back an acknowledgement and removes the entry comprising port identifier 165 of port 135 from database 142. Switch 103 runs state machine 250 based on database 142 and transitions redundant virtual link aggregation group 130 from active mode 270 to protection mode 280, as described in conjunction with FIG. 2B. As a result, the port status becomes “selected” for the entries with a “backup” port role and the port status becomes “standby” for the entries with an “active” port role in database 142.



FIG. 5D illustrates an exemplary data structure with selected active links in a redundant virtual link aggregation group in response to a failure recovery, in accordance with an embodiment of the present invention. This example shows database 142 in response to a recovery from failure 502, which makes ports 132 and 133 available again, and failure 506, which makes port 136 unavailable. Upon detecting a recovery from failure 502 to port 132, switch 103 sends a notification to switches 104 and 105 regarding the availability of the link coupled to port 132, and waits for an acknowledgement from switches 104 and 105. Upon receiving a respective acknowledgement, switch 103 adds an entry comprising port identifier 162 of port 132 to database 142. Switch 103 runs state machine 250 based on database 142 and determines that redundant virtual link aggregation group 130 remains in protection mode 280, as described in conjunction with FIG. 2B.


On the other hand, when switch 103 receives a notification from switch 104 regarding availability of port 133, switch 103 sends back an acknowledgement and adds an entry comprising port identifier 163 of port 133 to database 142. Switch 103 again runs state machine 250 based on database 142 and determines that redundant virtual link aggregation group 130 remains in protection mode 280, as described in conjunction with FIG. 2B. However, when switch 103 receives a notification from switch 105 regarding unavailability of port 135 due to failure 506, switch 103 removes the entry comprising port identifier 165 of port 135 from database 142. Switch 103 runs state machine 250 based on database 142 and transitions redundant virtual link aggregation group 130 from protection mode 280 to active mode 270, as described in conjunction with FIG. 2B. As a result, the port status becomes “selected” for the entries with an “active” port role and the port status becomes “standby” for the entries with a “backup” port role in database 142.


Recovery Operations



FIG. 6A presents a flowchart illustrating the process of a switch selecting local port status for a redundant virtual link aggregation group in response to a state change, in accordance with an embodiment of the present invention. During operation, the switch detects a state change of a local participant port of a redundant virtual link aggregation group (operation 602). The switch constructs a notification message indicating the state change of the local port in its payload (operation 604). This notification message can be a fabric control message of a fabric switch. The switch then sends the notification message to a respective other participant switch and sets a port flag of the local port to WAIT_ACK (operation 606). In some embodiments, sending a message includes identifying one or more local egress ports corresponding to the egress switch identifier of the message and transmitting the message via the identified port(s).


The switch then receives an acknowledgement (e.g., an acknowledgement message) from a remote participant switch (operation 608). The switch checks whether an acknowledgement from a respective remote participant switch has been received (operation 610). If not, the switch continues to receive an acknowledgement from a remote participant switch (operation 608). If the switch has received an acknowledgement from a respective remote participant switch, the switch sets the port flag to ALL_ACK (operation 612). The switch then checks the state change type (operation 614). If it is a “link down” state change (e.g., a link has become unavailable), the switch removes the entry comprising the port information of the local participant port from the local redundant virtual link aggregation group database (operation 616). If it is a “link up” state change (e.g., a link has become available), the switch adds an entry comprising the port information of the local participant port to the local database (operation 618).


Upon removing (operation 616) or adding (operation 618) the port information, the switch runs the state machine for the redundant virtual link aggregation group based on the database, as described in conjunction with FIG. 2A, and determines the mode for the redundant virtual link aggregation group (operation 620). The switch marks the port status of a respective entry in the database based on the determined mode of the redundant virtual link aggregation group (operation 622). The switch identifies the local port(s) with a “selected” port status and aggregates the links coupled to the identified port(s) in the redundant virtual link aggregation group (operation 624). In some embodiments, the switch sends a confirmation message a respective identified port to a respective remote participant switch (operation 626). This confirmation message indicates that the port is operational and ready to carry traffic for the redundant virtual link aggregation group. This confirmation message can be a fabric control message and can be sent for any link aggregation in a fabric switch.



FIG. 6B presents a flowchart illustrating the process of a switch selecting remote port status for a redundant virtual link aggregation group in response to a state change, in accordance with an embodiment of the present invention. During operation, the switch receives a notification message indicating a state change of a remote participant port of the redundant virtual link aggregation group from a remote participant switch (operation 652). The switch sends back an acknowledgement (operation 654) and checks the state change type (operation 656). If it is a “link down” state change (e.g., a link has become unavailable), the switch removes the entry comprising the port information of the remote port from the local redundant virtual link aggregation group database (operation 658). If it is a “link up” state change (e.g., a link has become available), the switch adds an entry comprising the port information of the remote port to the local database (operation 660).


Upon removing (operation 658) or adding (operation 660) the port information, the switch runs the state machine for the redundant virtual link aggregation group based on the database, as described in conjunction with FIG. 2A, and determines the mode for the redundant virtual link aggregation group (operation 662). The switch marks the port status of a respective entry in the database based on the determined mode of the redundant virtual link aggregation group (operation 664). In some embodiments, the switch receives a confirmation message from the remote participant switch indicating that the remote port is operational (operation 666). Upon receiving the confirmation message, the switch can mark this remote port to be ready to carry traffic for the redundant virtual link aggregation group (operation 668). This confirmation message can be a fabric control message and can be sent for any link aggregation in a fabric switch.


Forwarding



FIG. 7A presents a flowchart illustrating the process of a switch forwarding a packet received via an inter-switch port, in accordance with an embodiment of the present invention. During operation, the switch receives a fabric-encapsulated packet, which has a fabric-encapsulation header, with a virtual switch identifier as the egress switch identifier from a remote member switch via an inter-switch port (operation 702). This virtual switch identifier is associated with a redundant virtual link aggregation group. The fabric-encapsulated packet can be a TRILL or an IP packet and the virtual switch identifier can be a virtual RBridge identifier or a virtual IP address. The switch then checks whether the local switch is coupled to one or more selected link(s) (i.e., a link coupled to a participant port with a “selected” port role) (operation 704).


If the switch is not coupled to a selected link, the switch identifies the local inter-switch port(s) associated with the virtual switch identifier based on the local forwarding table (operation 706). In some embodiments, the switch maintains a mapping between the local inter-switch ports associated with remote participant switch(es) and the virtual switch identifier in the local forwarding table. If the switch has a plurality of remote participant switches with a “selected” port status (e.g., switch 103 in protection mode 280), the switch selects one of the identified local ports as the egress port for the fabric-encapsulated packet based on a switch selection policy (e.g., round robin, shortest distance, bandwidth, latency, hashing, etc.) (operation 708). The switch then transmits the fabric-encapsulated packet via the selected egress port (operation 710).


If the switch is coupled to a selected link, the switch decapsulates the fabric-encapsulation header and obtains the inner packet (e.g., an Ethernet frame) (operation 712). The switch identifies the local participant port(s) coupled to the corresponding selected link(s) based on the local forwarding table (operation 714). In some embodiments, the switch maintains a mapping between the local participant ports associated with the virtual switch identifier and the destination MAC address of the inner packet in the local forwarding table. If the switch has a plurality of such local ports (e.g., switch 103 in active mode 270), the switch selects one of the identified local ports as the egress port for the inner frame based on a port selection policy (e.g., round robin, bandwidth, hashing, etc.) (operation 716). The switch then transmits the inner packet via the selected egress port (operation 718).



FIG. 7B presents a flowchart illustrating the process of a switch forwarding a packet received via an edge port participating in a redundant virtual link aggregation group, in accordance with an embodiment of the present invention. During operation, the switch receives a packet (e.g., an Ethernet frame) via a local participant port in the redundant virtual link aggregation group (operation 752). The switch identifies a switch identifier associated with the destination address (e.g., the destination MAC address) of the received packet from the local forwarding table (operation 754). This switch identifier can be assigned to the switch from which the destination address is learned. The switch then encapsulates the received packet in a fabric-encapsulation header (operation 756) and assigns the identified switch identifier as the egress switch identifier of the encapsulation header (operation 758). The switch assigns the virtual switch identifier associated with the redundant virtual link aggregation group as the egress switch identifier of the encapsulation header (operation 760). The switch then sends the fabric-encapsulated packet to the next-hop switch associated with the identified switch identifier (operation 762). The switch can identify the next-hop switch from a local forwarding table which includes a next-hop switch for a respective member switch of a fabric switch.


Exemplary Switch



FIG. 8 illustrates an exemplary participant switch of a redundant virtual link aggregation group, in accordance with an embodiment of the present invention. Switch 800 includes a number of communication ports 802, a packet processor 810, a link aggregation module 830, a link management module 832, and a storage device 850. Switch 800 can also include switch modules (e.g., processing hardware of switch 800, such as its ASIC chips), which includes information based on which switch 800 processes packets (e.g., determines output ports for packets). Packet processor 810 extracts and processes header information from the received packets. Packet processor 810 can identify a switch identifier associated with the switch in the header of a packet.


In some embodiments, switch 800 maintains a membership in a fabric switch, as described in conjunction with FIG. 1. Switch 800 then includes a fabric switch module 820. Fabric switch module 820 maintains a configuration database in storage device 850 that maintains the configuration state of every switch within the fabric switch. Fabric switch module 820 maintains the state of the fabric switch, which is used to join other switches. Fabric switch module 820 can store configuration information associated with the fabric switch in a data structure in an object relational database 840 in storage device 850.


Communication ports 802 can include inter-switch communication channels for communication within the fabric switch. This inter-switch communication channel can be implemented via a regular communication port and based on any open or proprietary format. Communication ports 802 can also include one or more extension communication ports for communication between neighbor fabric switches. Communication ports 802 can include one or more TRILL ports capable of receiving frames encapsulated in a TRILL header. Communication ports 802 can also include one or more IP ports capable of receiving IP packets. An IP port is capable of receiving an IP packet and can be configured with an IP address. Packet processor 810 can process TRILL-encapsulated frames and/or IP packets.


During operation, link aggregation module 830 establishes a redundant virtual link aggregation group comprising a plurality of links coupled to switch 800 and one or more other participant switches. The plurality of links includes a first set of links coupling a first customer device and a second set of links coupling a second customer device, as described in conjunction with FIG. 1A. Link management module 832 determines a current mode of the redundant virtual link aggregation group, and operates the first set of links as active links carrying traffic for the redundant virtual link aggregation group and the second set of links as standby links for the first set of links. This determination is based on the current mode and a port role for one of the communication ports 802 which participates in the virtual link aggregation group.


Link management module 832 can determine the current mode by comparing a number of operational links in the first and second sets of links with a protection threshold value. In some embodiments, link management module 832 identifies an acknowledgment of a notification message from a remote participant switch. Upon receiving the acknowledgment from a respective remote participant switch, link management module 832 determines whether the first or second set of links is actively carrying traffic


In some embodiments, link aggregation module 830 maintains a redundant virtual link aggregation group database for the virtual link aggregation group, as described in conjunction with FIG. 2A. A respective entry in the database is associated with a participant port and includes a port role for the port. If the port role indicates that the port is coupled to an active link and the current mode indicates that the first set of links is actively carrying traffic, link management module 832 marks the entry as selected to carry traffic. On the other hand, if the port role indicates that the port is coupled to an active link and the current mode indicates that the second set of links is actively carrying traffic, link management module 832 marks the entry as standby. Furthermore, if the current mode indicates that the second set of links is currently active, link management module 832 operates the second set of links as active links carrying traffic for the redundant virtual link aggregation group.


In some embodiments, the virtual link aggregation group is represented as a virtual switch identifier associated with switch 800 and the other participant switches, as described in conjunction with FIG. 1B. Switch 800 then includes a forwarding module 870 which determines whether a local participant port is coupled to a link carrying traffic for the redundant virtual link aggregation group. If the port is coupled to a link carrying traffic, forwarding module 870 determines the port as an egress port for the inner packet of a fabric-encapsulated packet with the virtual switch identifier as the egress switch identifier. On the other hand, if none of the communication ports 802 is coupled to an operational link carrying traffic for the redundant virtual link aggregation group, forwarding module 870 determines an inter-switch port, which corresponds to another participant switch, as an egress port for the packet.


Note that the above-mentioned modules can be implemented in hardware as well as in software. In one embodiment, these modules can be embodied in computer-executable instructions stored in a memory which is coupled to one or more processors in switch 800. When executed, these instructions cause the processor(s) to perform the aforementioned functions.


In summary, embodiments of the present invention provide a switch and a method for facilitating a redundant virtual link aggregation group. In one embodiment, the switch includes a link aggregation module and a link management module. The link aggregation module establishes a virtual link aggregation group comprising a plurality of links coupled to the switch and one or more other switches. The plurality of links includes a first set of links coupling a first end device and a second set of links coupling a second end device. The link management module determines a current mode, which indicates which of the sets of links is currently active, of the virtual link aggregation group. The link management module operates the first set of links as active links carrying traffic for the virtual link aggregation group and the second set of links as standby links for the first set of links based on the current mode and a port role of a port participating in the virtual link aggregation group. The port role indicates whether the port is coupled to an active link or a backup link.


The methods and processes described herein can be embodied as code and/or data, which can be stored in a computer-readable non-transitory storage medium. When a computer system reads and executes the code and/or data stored on the computer-readable non-transitory storage medium, the computer system performs the methods and processes embodied as data structures and code and stored within the medium.


The methods and processes described herein can be executed by and/or included in hardware modules or apparatus. These modules or apparatus may include, but are not limited to, an application-specific integrated circuit (ASIC) chip, a field-programmable gate array (FPGA), a dedicated or shared processor that executes a particular software module or a piece of code at a particular time, and/or other programmable-logic devices now known or later developed. When the hardware modules or apparatus are activated, they perform the methods and processes included within them.


The foregoing descriptions of embodiments of the present invention have been presented only for purposes of illustration and description. They are not intended to be exhaustive or to limit this disclosure. Accordingly, many modifications and variations will be apparent to practitioners skilled in the art. The scope of the present invention is defined by the appended claims.

Claims
  • 1. A switch, comprising: link aggregation circuitry configured to establish a single virtual link aggregation group (VLAG) comprising a first set of links and a second set of links, wherein a first end device is reachable via the VLAG through the first set of links from the switch and a second switch in one hop, anda second end device is reachable via the VLAG through the second set of links from the switch and the second switch in one hop; andlink management circuitry configured to: determine a current mode of the VLAG by comparing a respective number of operational links in the first set of links and the second set of links with a protection threshold value, the current mode indicating which one of the first set of links and the second set of links is currently active in the VLAG; andoperate, based on the current mode and a port role, the first set of links as active links carrying traffic for the VLAG and the second set of links as standby links for the first set of links, the port role being of a respective port participating in the VLAG and the port role indicating whether the respective port corresponds to an active link or a standby link,wherein comparing the respective number of operational links includes determining whether the number of operational links in one of the first set of links and the second set of links is lower than the protection threshold value and whether the number of operational links in another of the first and second set of links is greater than or equal to the protection threshold value.
  • 2. The switch of claim 1, wherein the link management circuitry is further configured to: identify an acknowledgment message from the second switch indicating that the second switch has received port information associated with a port of the switch participating in the VLAG; andin response to identifying the acknowledgment message from the second switch, select the first set of links from the first and second sets of links for actively carrying traffic.
  • 3. The switch of claim 1, wherein the link aggregation circuitry is further configured to maintain a database for the VLAG, anda respective entry in the database is associated with a port participating the VLAG and includes a port role for the port.
  • 4. The switch of claim 3, wherein the link management circuitry is further configured to update an indicator in a first entry, which is associated with a first port in the first set of links, in the database to indicate that the first port is selected to carry traffic, andthe first port is selected to carry traffic in response to a port role of the first entry indicating that the first port corresponds to an active link and the current mode indicating that the first set of links is actively carrying traffic.
  • 5. The switch of claim 3, wherein the link management circuitry is further configured to update an indicator in a second entry, which is associated with a second port in the second set of links, in the database to indicate that the second port is a standby port, andthe second port operates as a standby port in response to a port role of the second port indicating that the second port corresponds to an active link and the current mode indicating that the first set of links is actively carrying traffic.
  • 6. The switch of claim 1, wherein the link management circuitry is further configured to operate the second set of links as active links carrying traffic for the VLAG and the first set of links as standby links for the second set of links in response to the current mode indicating that the second set of links is currently active.
  • 7. The switch of claim 1, wherein the VLAG is represented as a single virtual switch identifier, and wherein the virtual switch identifier is associated with the switch and the second switch.
  • 8. The switch of claim 1, further comprising forwarding circuitry configured to: determine whether a local port participating in the VLAG corresponds to an operational link in the first set of links; andin response to determining that the local port corresponds to an operational link in the first set of links, determine the local port as an egress port for a packet destined to the first end device.
  • 9. The switch of claim 8, wherein the forwarding circuitry is further configured to, in response to determining that no local port corresponds to an operational link in the first set of links, determine an inter-switch port as an egress port for the packet, wherein the second switch is reachable via the inter-switch port.
  • 10. The switch of claim 1, wherein the switch and the second switch are member switches of a network of interconnected switches, andthe network of interconnected switches is identified by a fabric identifier associated with a respective member of the network of interconnected switches.
  • 11. A method, comprising: establishing a single virtual link aggregation group (VLAG) comprising a first set of links and a second set of links, wherein a first end device is reachable via the VLAG through the first set of links from a switch and a second switch in one hop, anda second end device is reachable via the VLAG through the second set of links from the switch and the second switch in one hop;determining a current mode of the VLAG by comparing a respective number of operational links in the first set of links and the second set of links with a protection threshold value, the current mode indicating which one of the first set of links and die second set of links is currently active in the VLAG; andoperating, based on the current mode and a port role, the first set of links as active links carrying traffic for the VLAG and the second set of links as standby links for the first set of links, the port role being of a respective port participating in the VLAG and the port role indicating whether the respective port corresponds to an active link or a standby link,wherein comparing the respective number of operational links includes determining whether the number of operational links in one of the first and second sets of links is lower than the protection threshold value and whether the number of operational links in another of the first and second set of links is greater than or equal to the threshold value.
  • 12. The method of claim 11, further comprising: identifying an acknowledgment message from the second switch indicating that the second switch has received port information associated with a switch port of a switch participating in the VLAG; andin response to identifying the acknowledgment message from the second switch, selecting the first set of links from the first and second sets of links for actively carrying traffic.
  • 13. The method of claim 11, further comprising: maintaining a database for the VLAG, whereina respective entry in the database is associated with a port participating the VLAG and includes a port role for the port.
  • 14. The method of claim 13, further comprising: updating an indicator in a first entry, which is associated with a first port in the first set of links, in the database to indicate that the first port is selected to carry traffic, whereinthe first port is selected to carry traffic in response to a port role of the first entry indicating that the first port corresponds to an active link and the current mode indicating that the first set of links is actively carrying traffic.
  • 15. The method of claim 13, further comprising: updating an indicator in a second entry, which is associated with a second port in the second set of links, in the database to indicate that the second port is a standby port, whereinthe second port operate as a standby port in response to a port role of the second port indicating that the second port corresponds to an active link and the current mode indicating that the first set of links is actively carrying traffic.
  • 16. The method of claim 11, further comprising: operating the second set of links as active links carrying traffic for the VLAG and the first set of links as standby links for the second set of links in response to the current mode indicating that the second set of links is currently active.
  • 17. The method of claim 11, wherein the VLAG is represented as a single virtual switch identifier, andthe virtual switch identifier is associated with the switch and the second switch.
  • 18. The method of claim 11, further comprising: determining whether a port of the switch corresponds to an operational link in the first set of links; andin response to determining that the port corresponds to an operational link in the first set of links, determining the port as an egress port for a packet destined to the first end device.
  • 19. The method of claim 18, further comprising: in response to determining that no port of the switch corresponds to an operational link in the first set of links, determining an inter-switch port of the switch as an egress port for the packet, wherein the second switch is reachable via the inter-switch port.
  • 20. The method of claim 11, wherein the switch and the second switch are member switches of a network of interconnected switches, andthe network of interconnected switches is identified by a fabric identifier associated with a respective member of the network of interconnected switches.
  • 21. A non-transitory computer-readable storage medium storing instructions that when executed by a computing system cause the computing system to perform a method, the method comprising: establishing a single virtual link aggregation group (VLAG) comprising a first set of links and a second set of links, wherein a first end device is reachable via the VLAG through the first set of links from a switch and a second switch in one hop, anda second end device is reachable via the VLAG through the second set of links from the switch and the second switch in one hop;determining a current mode of the VLAG by comparing a respective number of operational links in the first set of links and the second set of links with a protection threshold value, the current mode indicating which one of the first set of links and die second set of links is currently active in the VLAG; andoperating, based on the current mode and a port role, the first set of links as active links carrying traffic for the VLAG and the second set of links as standby links for the first set of links, the port role being of a respective port participating in the VLAG and the port role indicating whether the respective port corresponds to an active link or a standby link,wherein comparing the respective number of operational links includes determining whether the number of operational links in one of the first set of links and the second set of links is lower than the protection threshold value and whether the number of operational links in another of the first and second set of links is greater than or equal to the threshold value.
RELATED APPLICATION

This application claims the benefit of U.S. Provisional Application No. 61/968,141, titled “Distributed Active-Passive redundancy in vLAG,” by inventors Shivalingayya Chikkamath, Prabu Thayalan, and Mythilikanth Raman, filed 20 Mar. 2014; and U.S. Provisional Application No. 61/975,625, titled “Distributed Active-Passive redundancy in vLAG,” by inventors Shivalingayya Chikkamath, Prabu Thayalan, and Mythilikanth Raman, filed 4 Apr. 2014, the disclosures of which are incorporated by reference herein. The present disclosure is related to U.S. Pat. No. 8,867,552, Ser. No. 13/087,239, titled “Virtual Cluster Switching,” by inventors Suresh Vobbilisetty and Dilip Chatwani, issued 21 Oct. 2014; and U.S. Pat. No. 8,665,886, Ser. No. 12/725,249, titled “Redundant Host Connection in a Routed Network,” by inventors Somesh Gupta, Anoop Ghanwani, Phanidhar Koganti, and Shunjia Yu, issued 4 Mar. 2014, the disclosures of which are incorporated by reference herein.

US Referenced Citations (685)
Number Name Date Kind
829529 Keathley Aug 1906 A
5390173 Spinney Feb 1995 A
5802278 Isfeld Sep 1998 A
5878232 Marimuthu Mar 1999 A
5879173 Poplawski Mar 1999 A
5959968 Chin Sep 1999 A
5973278 Wehrill, III Oct 1999 A
5983278 Chong Nov 1999 A
5995262 Hirota Nov 1999 A
6041042 Bussiere Mar 2000 A
6085238 Yuasa Jul 2000 A
6092062 Lohman Jul 2000 A
6104696 Kadambi Aug 2000 A
6122639 Babu Sep 2000 A
6185214 Schwartz Feb 2001 B1
6185241 Sun Feb 2001 B1
6295527 McCormack Sep 2001 B1
6331983 Haggerty Dec 2001 B1
6438106 Pillar Aug 2002 B1
6498781 Bass Dec 2002 B1
6542266 Phillips Apr 2003 B1
6553029 Alexander Apr 2003 B1
6571355 Linnell May 2003 B1
6583902 Yuen Jun 2003 B1
6633761 Singhal Oct 2003 B1
6636963 Stein Oct 2003 B1
6771610 Seaman Aug 2004 B1
6816462 Booth, III Nov 2004 B1
6870840 Hill Mar 2005 B1
6873602 Ambe Mar 2005 B1
6920503 Nanji Jul 2005 B1
6937576 DiBenedetto Aug 2005 B1
6956824 Mark Oct 2005 B2
6957269 Williams Oct 2005 B2
6975581 Medina Dec 2005 B1
6975864 Singhal Dec 2005 B2
7016352 Chow Mar 2006 B1
7061877 Gummalla Jun 2006 B1
7062177 Grivna Jun 2006 B1
7097308 Kim et al. Aug 2006 B2
7173934 Lapuh Feb 2007 B2
7197308 Singhal Mar 2007 B2
7206288 Cometto Apr 2007 B2
7274694 Cheng Sep 2007 B1
7310664 Merchant Dec 2007 B1
7313637 Tanaka Dec 2007 B2
7316031 Griffith Jan 2008 B2
7330897 Baldwin Feb 2008 B2
7380025 Riggins May 2008 B1
7397768 Betker Jul 2008 B1
7397794 Lacroute Jul 2008 B1
7430164 Bare Sep 2008 B2
7453888 Zabihi Nov 2008 B2
7477894 Sinha Jan 2009 B1
7480258 Shuen Jan 2009 B1
7508757 Ge Mar 2009 B2
7558195 Kuo Jul 2009 B1
7558273 Grosser Jul 2009 B1
7571447 Ally Aug 2009 B2
7599901 Mital Oct 2009 B2
7653056 Dianes Jan 2010 B1
7688960 Aubuchon Mar 2010 B1
7690040 Frattura Mar 2010 B2
7716370 Devarapalli May 2010 B1
7720076 Dobbins May 2010 B2
7729296 Choudhary Jun 2010 B1
7787480 Mehta Aug 2010 B1
7792920 Istvan Sep 2010 B2
7796593 Ghosh Sep 2010 B1
7801021 Triantafillis Sep 2010 B1
7808992 Homchaudhuri Oct 2010 B2
7836332 Hara Nov 2010 B2
7843906 Chidambaram et al. Nov 2010 B1
7843907 Abou-Emara Nov 2010 B1
7860097 Lovett Dec 2010 B1
7898959 Arad Mar 2011 B1
7912091 Krishnan Mar 2011 B1
7924837 Shabtay Apr 2011 B1
7937438 Miller May 2011 B1
7937756 Kay May 2011 B2
7945941 Sinha May 2011 B2
7949638 Goodson May 2011 B1
7957386 Aggarwal Jun 2011 B1
8018938 Fromm Sep 2011 B1
8027354 Portolani Sep 2011 B1
8054832 Shukla Nov 2011 B1
8068442 Kompella Nov 2011 B1
8078704 Lee Dec 2011 B2
8090805 Chawla Jan 2012 B1
8102781 Smith Jan 2012 B2
8102791 Tang Jan 2012 B2
8116307 Thesayi Feb 2012 B1
8125928 Mehta Feb 2012 B2
8134922 Elangovan Mar 2012 B2
8155150 Chung Apr 2012 B1
8160063 Maltz Apr 2012 B2
8160080 Arad Apr 2012 B1
8170038 Belanger May 2012 B2
8175107 Yalagandula May 2012 B1
8095774 Lambeth Jun 2012 B1
8194674 Pagel Jun 2012 B1
8195774 Lambeth Jun 2012 B2
8204061 Sane Jun 2012 B1
8213313 Doiron Jul 2012 B1
8213336 Smith Jul 2012 B2
8230069 Korupolu Jul 2012 B2
8239960 Frattura Aug 2012 B2
8249069 Raman Aug 2012 B2
8270401 Barnes Sep 2012 B1
8295291 Ramanathan Oct 2012 B1
8295921 Ramanathan Oct 2012 B2
8301686 Appajodu Oct 2012 B1
8339994 Gnanasekaran Dec 2012 B2
8351352 Eastlake Jan 2013 B1
8369335 Jha Feb 2013 B2
8369347 Xiong Feb 2013 B2
8392496 Linden Mar 2013 B2
8451717 Srikrishnan May 2013 B2
8462774 Page Jun 2013 B2
8465774 Page Jun 2013 B2
8520595 Yadav Aug 2013 B2
8553710 White Oct 2013 B1
8595479 Radhakrishnan Nov 2013 B2
8599850 Jha Dec 2013 B2
8599864 Chung Dec 2013 B2
8615008 Natarajan Dec 2013 B2
8619788 Sankaran Dec 2013 B1
8625616 Vobbilisetty Jan 2014 B2
8705526 Hasan Apr 2014 B1
8706905 McGlaughlin Apr 2014 B1
8717895 Koponen May 2014 B2
8724456 Hong May 2014 B1
8730963 Grosser, Jr. May 2014 B1
8792501 Rustagi Jul 2014 B1
8798045 Aybay Aug 2014 B1
8798055 An Aug 2014 B1
8804732 Hepting Aug 2014 B1
8804736 Drake Aug 2014 B1
8806031 Kondur Aug 2014 B1
8826385 Congdon Sep 2014 B2
8918631 Kumar Dec 2014 B1
8937865 Kumar Jan 2015 B1
8948181 Kapadia Feb 2015 B2
8971173 Choudhury Mar 2015 B1
9019976 Gupta Apr 2015 B2
9178793 Marlow Nov 2015 B1
9231890 Vobbilisetty Jan 2016 B2
9350680 Thayalan May 2016 B2
9401818 Venkatesh Jul 2016 B2
9438447 Basso Sep 2016 B2
9450870 Ananthapadmanabha Sep 2016 B2
9524173 Guntaka Dec 2016 B2
9626255 Guntaka Apr 2017 B2
9628407 Guntaka Apr 2017 B2
20010005527 Vaeth Jun 2001 A1
20010055274 Hegge Dec 2001 A1
20020021701 Lavian Feb 2002 A1
20020027885 Ben-Ami Mar 2002 A1
20020039350 Wang Apr 2002 A1
20020054593 Morohashi May 2002 A1
20020087723 Williams Jul 2002 A1
20020091795 Yip Jul 2002 A1
20020138628 Tingley Sep 2002 A1
20020161867 Cochran Oct 2002 A1
20030026290 Umayabashi Feb 2003 A1
20030041085 Sato Feb 2003 A1
20030093567 Lolayekar May 2003 A1
20030097464 Martinez May 2003 A1
20030097470 Lapuh May 2003 A1
20030123393 Feuerstraeter Jul 2003 A1
20030147385 Montalvo Aug 2003 A1
20030152075 Hawthorne Aug 2003 A1
20030174706 Shankar Sep 2003 A1
20030189905 Lee Oct 2003 A1
20030189930 Terrell Oct 2003 A1
20030208616 Laing Nov 2003 A1
20030216143 Roese Nov 2003 A1
20030223428 BlanquerGonzalez Dec 2003 A1
20030233534 Bernhard Dec 2003 A1
20040001433 Gram Jan 2004 A1
20040003094 See Jan 2004 A1
20040010600 Baldwin Jan 2004 A1
20040088668 Bornowski Jan 2004 A1
20040037295 Tanaka Feb 2004 A1
20040047349 Fujita Mar 2004 A1
20040049699 Griffith Mar 2004 A1
20040057430 Paavolainen Mar 2004 A1
20040081171 Finn Apr 2004 A1
20040088437 Stimac May 2004 A1
20040095900 Siegel May 2004 A1
20040117508 Shimizu Jun 2004 A1
20040120326 Yoon Jun 2004 A1
20040165595 Holmgren Aug 2004 A1
20040165596 Garcia Aug 2004 A1
20040205234 Barrack Oct 2004 A1
20040213232 Regan Oct 2004 A1
20040225725 Enomoto Nov 2004 A1
20040243673 Goyal Dec 2004 A1
20050007951 Lapuh Jan 2005 A1
20050025179 McLaggan Feb 2005 A1
20050036488 Kalkunte Feb 2005 A1
20050044199 Shiga Feb 2005 A1
20050074001 Mattes Apr 2005 A1
20050094568 Judd May 2005 A1
20050094630 Valdevit May 2005 A1
20050108375 Hallak-Stamler May 2005 A1
20050111352 Ho May 2005 A1
20050122979 Gross Jun 2005 A1
20050152335 Lodha Jul 2005 A1
20050157751 Rabie Jul 2005 A1
20050169188 Cometto Aug 2005 A1
20050195813 Ambe Sep 2005 A1
20050207423 Herbst Sep 2005 A1
20050213561 Yao Sep 2005 A1
20050220096 Friskney Oct 2005 A1
20050259586 Hafid Nov 2005 A1
20050265330 Suzuki Dec 2005 A1
20050265356 Kawarai Dec 2005 A1
20050278565 Frattura Dec 2005 A1
20060007869 Hirota Jan 2006 A1
20060018302 Ivaldi Jan 2006 A1
20060029055 Perera Feb 2006 A1
20060034292 Wakayama Feb 2006 A1
20060036648 Frey Feb 2006 A1
20060036765 Weyman Feb 2006 A1
20060039366 Ghosh Feb 2006 A1
20060059163 Frattura Mar 2006 A1
20060062187 Rune Mar 2006 A1
20060072550 Davis Apr 2006 A1
20060083172 Jordan Apr 2006 A1
20060083254 Ge Apr 2006 A1
20060093254 Ge Apr 2006 A1
20060092860 Higashitaniguchi May 2006 A1
20060098589 Kreeger May 2006 A1
20060126511 Youn Jun 2006 A1
20060140130 Kalkunte Jun 2006 A1
20060155828 Ikeda Jul 2006 A1
20060168109 Warmenhoven Jul 2006 A1
20060184937 Abels Aug 2006 A1
20060206655 Chappell Sep 2006 A1
20060209886 Silberman Sep 2006 A1
20060221960 Borgione Oct 2006 A1
20060227776 Chandrasekaran Oct 2006 A1
20060235995 Bhatia Oct 2006 A1
20060242311 Mai Oct 2006 A1
20060242398 Fontijn Oct 2006 A1
20060245439 Sajassi Nov 2006 A1
20060251067 Desanti Nov 2006 A1
20060256767 Suzuki Nov 2006 A1
20060265515 Shiga Nov 2006 A1
20060285499 Tzeng Dec 2006 A1
20060291480 Cho Dec 2006 A1
20060294413 Filz Dec 2006 A1
20070036178 Hares Feb 2007 A1
20070053294 Ho Mar 2007 A1
20070061817 Atkinson Mar 2007 A1
20070074052 Hemmah Mar 2007 A1
20070081530 Nomura Apr 2007 A1
20070083625 Chamdani Apr 2007 A1
20070086362 Kato Apr 2007 A1
20070094464 Sharma Apr 2007 A1
20070097968 Du May 2007 A1
20070098006 Parry May 2007 A1
20070110068 Sekiguchi May 2007 A1
20070116224 Burke May 2007 A1
20070116422 Reynolds May 2007 A1
20070121617 Kanekar May 2007 A1
20070130295 Rastogi Jun 2007 A1
20070156659 Lim Jul 2007 A1
20070177525 Wijnands Aug 2007 A1
20070177597 Ju Aug 2007 A1
20070183313 Narayanan Aug 2007 A1
20070183393 Boyd Aug 2007 A1
20070206762 Chandra Sep 2007 A1
20070211712 Fitch Sep 2007 A1
20070220059 Lu Sep 2007 A1
20070226214 Smits Sep 2007 A1
20070230472 Jesuraj Oct 2007 A1
20070238343 Velleca Oct 2007 A1
20070258449 Bennett Nov 2007 A1
20070274234 Kubota Nov 2007 A1
20070280223 Pan Dec 2007 A1
20070289017 Copeland, III Dec 2007 A1
20070297406 Rooholamini Dec 2007 A1
20080052487 Akahane Feb 2008 A1
20080056135 Lee Mar 2008 A1
20080056300 Williams Mar 2008 A1
20080057918 Abrant Mar 2008 A1
20080065760 Damm Mar 2008 A1
20080075078 Watanabe Mar 2008 A1
20080080517 Roy Apr 2008 A1
20080095160 Yadav Apr 2008 A1
20080101386 Gray May 2008 A1
20080112133 Torudbakken May 2008 A1
20080133760 Berkvens Jun 2008 A1
20080159260 Vobbilisetty Jul 2008 A1
20080159277 Vobbilisetty Jul 2008 A1
20080165705 Umayabashi Jul 2008 A1
20080172492 Raghunath Jul 2008 A1
20080181196 Regan Jul 2008 A1
20080186968 Farinacci Aug 2008 A1
20080186981 Seto Aug 2008 A1
20080205377 Chao Aug 2008 A1
20080219172 Mohan Sep 2008 A1
20080225852 Raszuk Sep 2008 A1
20080225853 Melman Sep 2008 A1
20080240129 Elmeleegy Oct 2008 A1
20080253380 Cazares Oct 2008 A1
20080267179 Lavigne Oct 2008 A1
20080279196 Friskney Nov 2008 A1
20080285458 Lysne Nov 2008 A1
20080285555 Ogasahara Nov 2008 A1
20080288020 Einav Nov 2008 A1
20080298248 Roeck Dec 2008 A1
20080304498 Jorgensen Dec 2008 A1
20080304519 Koenen Dec 2008 A1
20080310342 Kruys Dec 2008 A1
20090022069 Khan Jan 2009 A1
20090024734 Merbach Jan 2009 A1
20090037607 Farinacci Feb 2009 A1
20090037977 Gai Feb 2009 A1
20090041046 Hirata Feb 2009 A1
20090042270 Dolly Feb 2009 A1
20090044270 Shelly Feb 2009 A1
20090052326 Bergamasco Feb 2009 A1
20090067422 Poppe Mar 2009 A1
20090067442 Killian Mar 2009 A1
20090079560 Fries Mar 2009 A1
20090080345 Gray Mar 2009 A1
20090083445 Ganga Mar 2009 A1
20090092042 Yuhara Apr 2009 A1
20090092043 Lapuh Apr 2009 A1
20090094354 Rastogi Apr 2009 A1
20090106298 Furusho Apr 2009 A1
20090106405 Mazarick Apr 2009 A1
20090113408 Toeroe Apr 2009 A1
20090116381 Kanda May 2009 A1
20090122700 Aboba May 2009 A1
20090129384 Regan May 2009 A1
20090129389 Halna DeFretay May 2009 A1
20090138577 Casado May 2009 A1
20090138752 Graham May 2009 A1
20090144720 Roush Jun 2009 A1
20090161584 Guan Jun 2009 A1
20090161670 Shepherd Jun 2009 A1
20090168647 Holness Jul 2009 A1
20090199177 Edwards Aug 2009 A1
20090204965 Tanaka Aug 2009 A1
20090213783 Moreton Aug 2009 A1
20090213867 Devireddy Aug 2009 A1
20090222879 Kostal Sep 2009 A1
20090225752 Mitsumori Sep 2009 A1
20090232031 Vasseur Sep 2009 A1
20090245112 Okazaki Oct 2009 A1
20090245137 Hares Oct 2009 A1
20090245242 Carlson Oct 2009 A1
20090246137 Hadida Oct 2009 A1
20090249444 Macauley Oct 2009 A1
20090252049 Ludwig Oct 2009 A1
20090252061 Small Oct 2009 A1
20090252503 Ishigami Oct 2009 A1
20090260083 Szeto Oct 2009 A1
20090279558 Davis Nov 2009 A1
20090279701 Moisand Nov 2009 A1
20090292858 Lambeth Nov 2009 A1
20090323698 LeFaucheur Dec 2009 A1
20090323708 Ihle Dec 2009 A1
20090327392 Tripathi Dec 2009 A1
20090327462 Adams Dec 2009 A1
20100002382 Aybay Jan 2010 A1
20100027420 Smith Feb 2010 A1
20100027429 Jorgens Feb 2010 A1
20100042869 Szabo Feb 2010 A1
20100046471 Hattori Feb 2010 A1
20100054260 Pandey Mar 2010 A1
20100061269 Banerjee Mar 2010 A1
20100074175 Banks Mar 2010 A1
20100085981 Gupta Apr 2010 A1
20100097941 Carlson Apr 2010 A1
20100103813 Allan Apr 2010 A1
20100103939 Carlson Apr 2010 A1
20100114818 Lier May 2010 A1
20100131636 Suri May 2010 A1
20100157844 Casey Jun 2010 A1
20100158024 Sajassi Jun 2010 A1
20100165877 Shukla Jul 2010 A1
20100165995 Mehta Jul 2010 A1
20100168467 Johnston Jul 2010 A1
20100169467 Shukla Jul 2010 A1
20100169948 Budko Jul 2010 A1
20100182920 Matsuoka Jul 2010 A1
20100189119 Sawada Jul 2010 A1
20100192225 Ma Jul 2010 A1
20100195489 Zhou Aug 2010 A1
20100195529 Liu Aug 2010 A1
20100214913 Kompella Aug 2010 A1
20100215042 Sato Aug 2010 A1
20100215049 Raza Aug 2010 A1
20100220724 Rabie Sep 2010 A1
20100226368 Mack-Crane Sep 2010 A1
20100226381 Mehta Sep 2010 A1
20100246388 Gupta Sep 2010 A1
20100246580 Kaganoi Sep 2010 A1
20100254703 Kirkpatrick Oct 2010 A1
20100257263 Casado Oct 2010 A1
20100258263 Douxchamps Oct 2010 A1
20100265849 Harel Oct 2010 A1
20100271960 Krygowski Oct 2010 A1
20100272107 Papp Oct 2010 A1
20100281106 Ashwood-Smith Nov 2010 A1
20100284414 Agarwal Nov 2010 A1
20100284418 Gray Nov 2010 A1
20100284698 McColloch Nov 2010 A1
20100287262 Elzur Nov 2010 A1
20100287548 Zhou Nov 2010 A1
20100290464 Assarpour Nov 2010 A1
20100290472 Raman Nov 2010 A1
20100290473 Enduri Nov 2010 A1
20100299527 Arunan Nov 2010 A1
20100303071 Kotalwar Dec 2010 A1
20100303075 Tripathi Dec 2010 A1
20100303083 Belanger Dec 2010 A1
20100309820 Rajagopalan Dec 2010 A1
20100309912 Mehta Dec 2010 A1
20100316055 Belanger Dec 2010 A1
20100329110 Rose Dec 2010 A1
20100329265 Lapuh Dec 2010 A1
20110007738 Berman Jan 2011 A1
20110019678 Mehta Jan 2011 A1
20110032945 Mullooly Feb 2011 A1
20110035489 McDaniel Feb 2011 A1
20110035498 Shah Feb 2011 A1
20110044339 Kotalwar Feb 2011 A1
20110044352 Chaitou Feb 2011 A1
20110051723 Rabie Mar 2011 A1
20110058547 Waldrop Mar 2011 A1
20110064086 Xiong Mar 2011 A1
20110064089 Hidaka Mar 2011 A1
20110072208 Gulati Mar 2011 A1
20110085557 Gnanasekaran Apr 2011 A1
20110085560 Chawla Apr 2011 A1
20110085562 Bao Apr 2011 A1
20110085563 Kotha Apr 2011 A1
20110088011 Ouali Apr 2011 A1
20110110266 Li May 2011 A1
20110134802 Rajagopalan Jun 2011 A1
20110134803 Dalvi Jun 2011 A1
20110134924 Hewson Jun 2011 A1
20110134925 Safrai Jun 2011 A1
20110142053 Van Jun 2011 A1
20110142062 Wang Jun 2011 A1
20110149526 Turner Jun 2011 A1
20110158113 Nanda Jun 2011 A1
20110161494 McDysan Jun 2011 A1
20110161695 Okita Jun 2011 A1
20110176412 Stine Jul 2011 A1
20110188373 Saito Aug 2011 A1
20110194403 Sajassi Aug 2011 A1
20110194563 Shen Aug 2011 A1
20110225540 d'Entremont Sep 2011 A1
20110228767 Singla Sep 2011 A1
20110228780 Ashwood-Smith Sep 2011 A1
20110231570 Altekar Sep 2011 A1
20110231574 Saunderson Sep 2011 A1
20110235523 Jha Sep 2011 A1
20110243133 Villait Oct 2011 A9
20110243136 Raman Oct 2011 A1
20110246669 Kanada Oct 2011 A1
20110255538 Srinivasan Oct 2011 A1
20110255540 Mizrahi Oct 2011 A1
20110261828 Smith Oct 2011 A1
20110268118 Schlansker Nov 2011 A1
20110268120 Vobbilisetty Nov 2011 A1
20110268125 Vobbilisetty Nov 2011 A1
20110273988 Tourrilhes Nov 2011 A1
20110273990 Rajagopalan Nov 2011 A1
20110274114 Dhar Nov 2011 A1
20110280572 Vobbilisetty Nov 2011 A1
20110286357 Haris Nov 2011 A1
20110286457 Ee Nov 2011 A1
20110286462 Kompella Nov 2011 A1
20110055274 Hegge Dec 2011 A1
20110292947 Vobbilisetty Dec 2011 A1
20110296052 Guo Dec 2011 A1
20110299391 Vobbilisetty Dec 2011 A1
20110299413 Chatwani Dec 2011 A1
20110299414 Yu Dec 2011 A1
20110299527 Yu Dec 2011 A1
20110299528 Yu Dec 2011 A1
20110299531 Yu Dec 2011 A1
20110299532 Yu Dec 2011 A1
20110299533 Yu Dec 2011 A1
20110299534 Koganti Dec 2011 A1
20110299535 Vobbilisetty Dec 2011 A1
20110299536 Cheng Dec 2011 A1
20110317559 Kern Dec 2011 A1
20120011240 Hara Jan 2012 A1
20120014261 Salam Jan 2012 A1
20120014387 Dunbar Jan 2012 A1
20120020220 Sugita Jan 2012 A1
20120027017 Rai Feb 2012 A1
20120033663 Guichard Feb 2012 A1
20120033665 Jacob Feb 2012 A1
20120033668 Humphries Feb 2012 A1
20120033669 Mohandas Feb 2012 A1
20120033672 Page Feb 2012 A1
20120039163 Nakajima Feb 2012 A1
20120042095 Kotha Feb 2012 A1
20120063363 Li Mar 2012 A1
20120075991 Sugita Mar 2012 A1
20120099567 Hart Apr 2012 A1
20120099602 Nagapudi Apr 2012 A1
20120099863 Xu Apr 2012 A1
20120102160 Breh Apr 2012 A1
20120106339 Mishra May 2012 A1
20120117438 Shaffer May 2012 A1
20120131097 Baykal May 2012 A1
20120131289 Taguchi May 2012 A1
20120134266 Roitshtein May 2012 A1
20120136999 Roitshtein May 2012 A1
20120147740 Nakash Jun 2012 A1
20120158997 Hsu Jun 2012 A1
20120163164 Terry Jun 2012 A1
20120170491 Kern Jul 2012 A1
20120177039 Berman Jul 2012 A1
20120210416 Mihelich Aug 2012 A1
20120221636 Surtani Aug 2012 A1
20120230225 Matthews Sep 2012 A1
20120239918 Huang Sep 2012 A1
20120243359 Keesara Sep 2012 A1
20120243539 Keesara Sep 2012 A1
20120250502 Brolin Oct 2012 A1
20120260079 Mruthyunjaya Oct 2012 A1
20120275297 Subramanian Nov 2012 A1
20120275347 Banerjee Nov 2012 A1
20120278804 Narayanasamy Nov 2012 A1
20120281700 Koganti Nov 2012 A1
20120287785 Kamble Nov 2012 A1
20120294192 Masood Nov 2012 A1
20120294194 Balasubramanian Nov 2012 A1
20120230800 Kamble Dec 2012 A1
20120320739 Kamath Dec 2012 A1
20120320800 Kamble Dec 2012 A1
20120327766 Tsai et al. Dec 2012 A1
20130003535 Sarwar Jan 2013 A1
20130003549 Matthews Jan 2013 A1
20130003608 Lei Jan 2013 A1
20130003737 Sinicrope Jan 2013 A1
20130003738 Koganti Jan 2013 A1
20130003747 Raman Jan 2013 A1
20130016606 Cirkovic Jan 2013 A1
20130028072 Addanki Jan 2013 A1
20130034015 Jaiswal Feb 2013 A1
20130034021 Jaiswal Feb 2013 A1
20130034094 Cardona Feb 2013 A1
20130044629 Biswas Feb 2013 A1
20130058354 Casado Mar 2013 A1
20130066947 Ahmad Mar 2013 A1
20130067466 Combs Mar 2013 A1
20130070762 Adams Mar 2013 A1
20130083693 Himura Apr 2013 A1
20130097345 Munoz Apr 2013 A1
20130114595 Mack-Crane et al. May 2013 A1
20130121142 Bai May 2013 A1
20130124707 Ananthapadmanabha May 2013 A1
20130124750 Anumala May 2013 A1
20130127848 Joshi May 2013 A1
20130132296 Talagala May 2013 A1
20130135811 Dunwoody May 2013 A1
20130136123 Ge May 2013 A1
20130145008 Kannan Jun 2013 A1
20130148546 Eisenhauer Jun 2013 A1
20130148663 Xiong Jun 2013 A1
20130156425 Kirkpatrick Jun 2013 A1
20130163591 Shukla Jun 2013 A1
20130194914 Agarwal Aug 2013 A1
20130201992 Masaki Aug 2013 A1
20130215754 Tripathi Aug 2013 A1
20130219473 Schaefer Aug 2013 A1
20130223221 Xu Aug 2013 A1
20130223438 Tripathi Aug 2013 A1
20130223449 Koganti Aug 2013 A1
20130238802 Sarikaya Sep 2013 A1
20130250289 Kettler Sep 2013 A1
20130250947 Zheng Sep 2013 A1
20130250951 Koganti Sep 2013 A1
20130250958 Watanabe Sep 2013 A1
20130259037 Natarajan Oct 2013 A1
20130259047 Bhagavathiperumal Oct 2013 A1
20130266015 Qu Oct 2013 A1
20130268590 Mahadevan Oct 2013 A1
20130272135 Leong Oct 2013 A1
20130294451 Li Nov 2013 A1
20130297757 Han Nov 2013 A1
20130301425 Chandra Nov 2013 A1
20130301642 Radhakrishnan Nov 2013 A1
20130308492 Baphna Nov 2013 A1
20130308641 Ackley Nov 2013 A1
20130308647 Rosset Nov 2013 A1
20130315125 Ravishankar Nov 2013 A1
20130315246 Zhang Nov 2013 A1
20130315586 Kipp Nov 2013 A1
20130322427 Stiekes Dec 2013 A1
20130329605 Nakil Dec 2013 A1
20130332660 Talagala Dec 2013 A1
20130336104 Talla Dec 2013 A1
20130346583 Low Dec 2013 A1
20140010239 Xu Jan 2014 A1
20140013324 Zhang Jan 2014 A1
20140019608 Kawakami Jan 2014 A1
20140019639 Ueno Jan 2014 A1
20140025736 Wang Jan 2014 A1
20140029412 Janardhanan Jan 2014 A1
20140029419 Jain Jan 2014 A1
20140044126 Sabhanatarajan Feb 2014 A1
20140050223 Foo Feb 2014 A1
20140056298 Vobbilisetty Feb 2014 A1
20140059225 Gasparakis Feb 2014 A1
20140064056 Sakata Mar 2014 A1
20140071987 Janardhanan Mar 2014 A1
20140086253 Yong Mar 2014 A1
20140092738 Grandhi Apr 2014 A1
20140105034 Huawei Apr 2014 A1
20140112122 Kapadia Apr 2014 A1
20140140243 Ashwood-Smith May 2014 A1
20140157251 Hocker Jun 2014 A1
20140169368 Grover Jun 2014 A1
20140188996 Lie Jul 2014 A1
20140192804 Ghanwani Jul 2014 A1
20140195694 Sait Jul 2014 A1
20140195695 Okita Jul 2014 A1
20140198656 Venkatesh Jul 2014 A1
20140211607 Li Jul 2014 A1
20140241147 Rajagopalan Aug 2014 A1
20140258446 Bursell Sep 2014 A1
20140269701 Kaushik Sep 2014 A1
20140269709 Benny Sep 2014 A1
20140269720 Arvindinivasan Sep 2014 A1
20140269733 Venkatesh Sep 2014 A1
20140298091 Carlen Oct 2014 A1
20140348168 Singh Nov 2014 A1
20140355477 Moopath Dec 2014 A1
20140362854 Addanki Dec 2014 A1
20140362859 Addanki Dec 2014 A1
20140376550 Khan Dec 2014 A1
20150009992 Zhang Jan 2015 A1
20150010007 Matsuhira Jan 2015 A1
20150016300 Devireddy Jan 2015 A1
20150030031 Zhou Jan 2015 A1
20150092593 Kompella Apr 2015 A1
20150103826 Davis Apr 2015 A1
20150110105 Kapur Apr 2015 A1
20150110111 Song Apr 2015 A1
20150110487 Fenkes Apr 2015 A1
20150117256 Sabaa Apr 2015 A1
20150117454 Koponen Apr 2015 A1
20150124837 Saltsidis May 2015 A1
20150127618 Alberti May 2015 A1
20150139037 Li May 2015 A1
20150139234 Hu May 2015 A1
20150143369 Zheng May 2015 A1
20150172098 Agarwal Jun 2015 A1
20150188753 Anumala Jul 2015 A1
20150188770 Naiksatam Jul 2015 A1
20150195093 Ramasubramani Jul 2015 A1
20150222506 Kizhakkiniyil Aug 2015 A1
20150248298 Gavrilov Sep 2015 A1
20150263897 Ganichev Sep 2015 A1
20150263899 Tubaltsev Sep 2015 A1
20150263991 MacChiano Sep 2015 A1
20150271088 Ao Sep 2015 A1
20150281066 Koley Oct 2015 A1
20150301901 Rath Oct 2015 A1
20150347468 Bester Dec 2015 A1
20150365320 Xu Dec 2015 A1
20160072899 Tung Mar 2016 A1
20160087885 Tripathi Mar 2016 A1
20160139939 Bosch May 2016 A1
20160182458 Shatzkamer Jun 2016 A1
20160212040 Bhagavathiperumal Jul 2016 A1
20160344640 Derlund Nov 2016 A1
20170012880 Yang Jan 2017 A1
20170026197 Venkatesh Jan 2017 A1
20170097841 Chang Apr 2017 A1
20180013614 Vobbilisetty Jan 2018 A1
Foreign Referenced Citations (38)
Number Date Country
1735062 Feb 2006 CN
1777149 May 2006 CN
101064682 Oct 2007 CN
101459618 Jun 2009 CN
101471899 Jul 2009 CN
101548511 Sep 2009 CN
101645880 Feb 2010 CN
102098237 Jun 2011 CN
102148749 Aug 2011 CN
102301663 Dec 2011 CN
102349268 Feb 2012 CN
102378176 Mar 2012 CN
102404181 Apr 2012 CN
102415065 Apr 2012 CN
102415065 Apr 2012 CN
102801599 Nov 2012 CN
102801599 Nov 2012 CN
102088388 Apr 2014 CN
0579567 May 1993 EP
0579567 Jan 1994 EP
0993156 Apr 2000 EP
0993156 Dec 2000 EP
1398920 Mar 2004 EP
1398920 Mar 2004 EP
2001167 Aug 2007 EP
1916807 Apr 2008 EP
2854352 Apr 2015 EP
2874359 May 2015 EP
2008056838 May 2008 WO
2009042919 Apr 2009 WO
2010111142 Sep 2010 WO
2010111142 Sep 2010 WO
2011132568 Oct 2011 WO
2011140028 Nov 2011 WO
2011140028 Nov 2011 WO
2012033663 Mar 2012 WO
2012093429 Jul 2012 WO
2014031781 Feb 2014 WO
Non-Patent Literature Citations (248)
Entry
Mahalingam “VXLAN: A Framework for Overlaying Virtualized Layer 2 Networks over Layer 3 Networks” Oct. 17, 2013 pp. 1-22, Sections 1, 4 and 4.1.
Office action dated Apr. 30, 2015, U.S. Appl. No. 13/351,513, filed Jan. 17, 2012.
Office Action dated Apr. 1, 2015, U.S. Appl. No. 13/656,438, filed Oct. 19, 2012.
Office Action dated May 21, 2015, U.S. Appl. No. 13/288,822, filed Nov. 3, 2011.
Siamak Azodolmolky et al. “Cloud computing networking: Challenges and opportunities for innovations”, IEEE Communications Magazine, vol. 51, No. 7, Jul. 1, 2013.
Office action dated Jun. 8, 2015, U.S. Appl. No. 14/178,042, filed Feb. 11, 2014.
Office Action dated Jun. 10, 2015, U.S. Appl. No. 13/890,150, filed May 8, 2013.
Office Action dated Jun. 18, 2015, U.S. Appl. No. 13/098,490, filed May 2, 2011.
Office Action dated Jun. 16, 2015, U.S. Appl. No. 13/048,817, filed Mar. 15, 2011.
Open Flow Configuration and Management Protocol 1.0 (OF-Config 1.0) Dec. 23, 2011.
Open Flow Switch Specification Version 1.2 Dec. 5, 2011.
Office action dated Feb. 2, 2016, U.S. Appl. No. 13/092,460, filed Apr. 22, 2011.
Office Action dated Feb. 2, 2016. U.S. Appl. No. 14/154,106, filed Jan. 13, 2014.
Office Action dated Feb. 3, 2016, U.S. Appl. No. 13/098,490, filed May 2, 2011.
Office Action dated Feb. 4, 2016, U.S. Appl. No. 13/557,105, filed Jul. 24, 2012.
Office Action dated Feb. 11, 2016, U.S. Appl. No. 14/488,173, filed Sep. 16, 2014.
Office Action dated Feb. 24, 2016, U.S. Appl. No. 13/971,397, filed Aug. 20, 2013.
Office Action dated Feb. 24, 2016, U.S. Appl. No. 12/705,508, filed Feb. 12, 2010.
Office Action dated Jul. 6, 2016, U.S. Appl. No. 14/618,941, filed Feb. 10, 2015.
Office Action dated Jul. 20, 2016, U.S. Appl. No. 14/510,913, filed Oct. 9, 2014.
Office Action dated Jul. 29, 2016, U.S. Appl. No. 14/473,941, filed Aug. 29, 2014.
Office Action dated Jul. 28, 2016, U.S. Appl. No. 14/284,212, filed May 21, 2016.
Eastlake, D. et al., ‘RBridges: TRILL Header Options’, Dec. 24, 2009, pp. 1-17, TRILL Working Group.
Perlman, Radia et al., ‘RBridge VLAN Mapping’, TRILL Working Group, Dec. 4, 2009, pp. 1-12.
Touch, J. et al., ‘Transparent Interconnection of Lots of Links (TRILL): Problem and Applicability Statement’, May 2009, Network Working Group, pp. 1-17.
‘RBridges: Base Protocol Specification’, IETF Draft, Perlman et al., Jun. 26, 2009.
Switched Virtual Networks. ‘Internetworking Moves Beyond Bridges and Routers’ Data Communications, McGraw Hill. New York, US, vol. 23, No. 12, Sep. 1, 1994 (Sep. 1, 1994), pp. 66-70,72,74, XP000462385 ISSN: 0363-6399.
Office Action dated Apr. 9, 2014, U.S. Appl. No. 13/092,724, filed Apr. 22, 2011.
Office action dated Aug. 29, 2014, U.S. Appl. No. 13/042,259, filed Mar. 7, 2011.
Office Action dated Mar. 14, 2014,U.S. Appl. No. 13/092,460, filed Apr. 22, 2011.
Office Action dated Mar. 26, 2014, U.S. Appl. No. 13/092,701, filed Apr. 22, 2011.
Office Action dated Apr. 9, 2014, U.S. Appl. No. 13/092,752, filed Apr. 22, 2011.
Office action dated Aug. 21, 2014, U.S. Appl. No. 13/184,526, filed Jul. 16, 2011.
Office Action dated Mar. 6, 2014, U.S. Appl. No. 13/425,238, filed Mar. 20, 2012.
Office Action dated Jun. 18, 2014, U.S. Appl. No. 13/440,861, filed Apr. 5, 2012.
Office Action dated Feb. 28, 2014, U.S. Appl. No. 13/351,513, filed Jan. 17, 2012.
Office Action dated May 9, 2014, U.S. Appl. No. 13/484,072, filed May 30, 2012.
Office Action dated May 14, 2014, U.S. Appl. No. 13/533,843, filed Jun. 26, 2012.
Office Action dated Feb. 20, 2014, U.S. Appl. No. 13/598,204, filed Aug. 29, 2012.
Office Action dated Jun. 6, 2014, U.S. Appl. No. 13/669,357, filed Nov. 5, 2012.
Brocade, ‘Brocade Fabrics OS (FOS) 6.2 Virtual Fabrics Feature Frequently Asked Questions’, pp. 1-6, 2009 Brocade Communications Systems, Inc.
Brocade, ‘FastIron and TurboIron 24x Configuration Guide’, Feb. 16, 2010.
Brocade, ‘The Effortless Network: Hyperedge Technology for the Campus LAN’ 2012.
Brocade ‘An Introduction to Brocade VCS Fabric Technology’, Dec. 3, 2012.
Christensen, M. et al., ‘Considerations for Internet Group Management Protocol (IGMP) and Multicast Listener Discovery (MLD) Snooping Switches’, May 2006.
FastIron Configuration Guide Supporting Ironware Software Release 07.0.00, Dec. 18, 2009.
Foundary FastIron Configuration Guide, Software Release FSX 04.2.00b, Software Release FWS 04.3.00, Software Release FGS 05.0.00a, Sep. 2008.
Huang, Nen-Fu et al., ‘An Effective Spanning Tree Algorithm for a Bridged LAN’, Mar. 16, 1992.
Knight, ‘Network Based IP VPN Architecture using Virtual Routers’, May 2003.
Knight P et al: ‘Layer 2 and 3 Virtual Private Networks: Taxonomy, Technology, and Standardization Efforts’, IEEE, Communications Magazine; IEEE Service Center, Piscataway, US, vol. 42, No. 6, Jun. 1, 2004 (Jun. 1, 2004), pp. 124-131, XP001198207, ISSN: 0163-6804, DOI: 1031109/MCOM.200431304248.
Knight S et al: ‘Virtual Router Redundancy Protocol’ Internet Citation Apr. 1, 1998 (Apr. 1, 1998), XP002135272 Retrieved from the Internet: URL:ftp://ftp.isi.edu/in-notes/rfc2338.txt [retrieved on Apr. 10, 2000].
Lapuh, Roger et al., ‘Split Multi-link Trunking (SMLT)’, draft-lapuh-network-smlt-08, Jul. 2008.
Lapuh, Roger et al., ‘Split Multi-Link Trunking (SMLT)’, Network Working Group, Oct. 2012.
Louati, Wajdi et al., ‘Network-based virtual personal overlay networks using programmable virtual routers’, IEEE Communications Magazine, Jul. 2005.
Narten, T. et al., ‘Problem Statement: Overlays for Network Virtualization d raft-narten-n vo3-over l ay-problem-statement-01’, Oct. 31, 2011.
Office Action for U.S. Appl. No. 13/042,259, filed Mar. 7, 2011, from Jaroenchonwanit, Bunjob, dated Jan. 16, 2014.
Office Action for U.S. Appl. No. 13/092,752, filed Apr. 22, 2011, from Park, Jung H., dated Jul. 18, 2013.
Office Action for U.S. Appl. No. 13/365,993, filed Feb. 3, 2012, from Cho, Hong Sol., dated Jul. 23, 2013.
Office Action for U.S. Appl. No. 12/725,249, filed Mar. 16, 2010, dated Apr. 26, 2013.
Office Action for U.S. Appl. No. 12/725,249, filed Mar. 16, 2010, dated Sep. 12, 2012.
Office Action for U.S. Appl. No. 12/950,968, filed Nov. 19, 2010, dated Jan. 4, 2013.
Office Action for U.S. Appl. No. 12/950,968, filed Nov. 19, 2010, dated Jun. 7, 2012.
Office Action for U.S. Appl. No. 12/950,974, filed Nov. 19, 2010, dated Dec. 20, 2012.
Office Action for U.S. Appl. No. 12/950,974, filed Nov. 19, 2010, dated May 24, 2012.
Office Action for U.S. Appl. No. 13/030,688, filed Feb. 18, 2011, dated Apr. 25, 2013.
Office Action for U.S. Appl. No. 13/030,806, filed Feb. 18, 2011, dated Dec. 3, 2012.
Office Action for U.S. Appl. No. 13/030,806, filed Feb. 18, 2011, dated Jun. 11, 2013.
Office Action for U.S. Appl. No. 13/042,259, filed Mar. 7, 2011, dated Mar. 18, 2013.
Office Action for U.S. Appl. No. 13/042,259, filed Mar. 7, 2011, dated Jul. 31, 2013.
Office Action for U.S. Appl. No. 13/044,301, filed Mar. 9, 2011, dated Feb. 22, 2013.
Office Action for U.S. Appl. No. 13/044,301, filed Mar. 9, 2011, dated Jun. 11, 2013.
Office Action for U.S. Appl. No. 13/044,326, filed Mar. 9, 2011, dated Oct. 2, 2013.
Office Action for U.S. Appl. No. 13/050,102, filed Mar. 17, 2011, dated Oct. 26, 2012.
Office Action for U.S. Appl. No. 13/050,102, filed Mar. 17, 2011, dated May 16, 2013.
Office Action for U.S. Appl. No. 13/087,239, filed Apr. 14, 2011, dated May 22, 2013.
Office Action for U.S. Appl. No. 13/092,460, filed Apr. 22, 2011, dated Jun. 21, 2013.
Office Action for U.S. Appl. No. 13/092,580, filed Apr. 22, 2011, dated Jun. 10, 2013.
Office Action for U.S. Appl. No. 13/092,701, filed Apr. 22, 2011, dated Jan. 28, 2013.
Office Action for U.S. Appl. No. 13/092,701, filed Apr. 22, 2011, dated Jul. 3, 2013.
Office Action for U.S. Appl. No. 13/092,724, filed Apr. 22, 2011, dated Feb. 5, 2013.
Office Action for U.S. Appl. No. 13/092,724, filed Apr. 22, 2011, dated Jul. 16, 2013.
Office Action for U.S. Appl. No. 13/092,752, filed Apr. 22, 2011, dated Feb. 5, 2013.
Office Action for U.S. Appl. No. 13/092,864, filed Apr. 22, 2011, dated Sep. 19, 2012.
Office Action for U.S. Appl. No. 13/092,873, filed Apr. 22, 2011, dated Jun. 19, 2013.
Office Action for U.S. Appl. No. 13/092,877, filed Apr. 22, 2011, dated Mar. 4, 2013.
Office Action for U.S. Appl. No. 13/092,877, filed Apr. 22, 2011, dated Sep. 5, 2013.
Office Action for U.S. Appl. No. 13/098,360, filed Apr. 29, 2011, dated May 31, 2013.
Office Action for U.S. Appl. No. 13/098,490, filed May 2, 2011, dated Dec. 21, 2012.
Office Action for U.S. Appl. No. 13/098,490, filed May 2, 2011, dated Mar. 27, 2014.
Office Action for U.S. Appl. No. 13/098,490, filed May 2, 2011, dated Jul. 9, 2013.
Office Action for U.S. Appl. No. 13/184,526, filed Jul. 16, 2011, dated Jan. 28, 2013.
Office Action for U.S. Appl. No. 13/184,526, filed Jul. 16, 2011, dated May 22, 2013.
Office Action for U.S. Appl. No. 13/312,903, filed Dec. 6, 2011, dated Jun. 13, 2013.
Office Action for U.S. Appl. No. 13/092,873, filed Apr. 22, 2011, dated Nov. 29, 2013.
Office Action for U.S. Appl. No. 13/184,526, filed Jul. 16, 2011, dated Dec. 2, 2013.
Office Action for U.S. Appl. No. 13/533,843, filed Jun. 26, 2012, dated Oct. 21, 2013.
Office Action for U.S. Appl. No. 13/598,204, filed Aug. 29, 2012, dated Feb. 20, 2014.
Perlman, Radia et al., ‘Challenges and Opportunities in the Design of TRILL: a Routed layer 2 Technology’, 2009.
Perlman, Radia et al., ‘RBridges: Base Protocol Specification; Draft-ietf-trill-rbridge-protocol-16.txt’, Mar. 3, 2010, pp. 1-117.
S. Nadas et al., ‘Virtual Router Redundancy Protocol (VRRP) Version 3 for IPv4 and IPv6’, Internet Engineering Task Force, Mar. 2010.
Zhai F. Hu et al. ‘RBridge: Pseudo-Nickname; draft-hu-trill-pseudonode-nickname-02.txt’, May 15, 2012.
‘An Introduction to Brocade VCS Fabric Technology’, BROCADE white paper, http://community.brocade.com/docs/DOC-2954, Dec. 3, 2012.
U.S. Appl. No. 13/030,806 Office Action dated Dec. 3, 2012.
Office action dated Jan. 10, 2014, U.S. Appl. No. 13/092,580, filed Apr. 22, 2011.
Office action dated Jan. 16, 2014, U.S. Appl. No. 13/042,259, filed Mar. 7, 2011.
Office action dated Jul. 31, 2013, U.S. Appl. No. 13/042,259, filed Mar. 7, 2011.
Office action dated Jan. 6, 2014, U.S. Appl. No. 13/092,877, filed Apr. 22, 2011.
Office action dated Oct. 2, 2013, U.S. Appl. No. 13/044,326, filed Mar. 9, 2011.
Office action dated Dec. 2, 2013, U.S. Appl. No. 13/184,526, filed Jul. 16, 2011.
Office action dated Nov. 29, 2013, U.S. Appl. No. 13/092,873, filed Apr. 22, 2011.
Office action dated Nov. 12, 2013, U.S. Appl. No. 13/312,903, filed Dec. 6, 2011.
Brocade ‘Brocade Unveils ‘The Effortless Network’’, http://newsroom.brocade.com/press-releases/brocade-unveils-the-effortless-network-nasdaq-brcd-0859535, 2012.
Kreeger, L. et al., ‘Network Virtualization Overlay Control Protocol Requirements draft-kreeger-nvo3-overlay-cp-00’, Jan. 30, 2012.
Office Action for U.S. Appl. No. 13/365,808, filed Jul. 18, 2013, dated Jul. 18, 2013.
Office Action for U.S. Appl. No. 13/092,887 dated Jan. 6, 2014.
Office action dated Apr. 26, 2012, U.S. Appl. No. 12/725,249, filed Mar. 16, 2010.
Office action dated Sep. 12, 2012, U.S. Appl. No. 12/725,249, filed Mar. 16, 2010.
Office action dated Dec. 21, 2012, U.S. Appl. No. 13/098,490, filed May 2, 2011.
Office action dated Mar. 27, 2014, U.S. Appl. No. 13/098,490, filed May 2, 2011.
Office action dated Jul. 9, 2013, U.S. Appl. No. 13/098,490, filed May 2, 2011.
Office action dated May 22, 2013, U.S. Appl. No. 13/087,239, filed Apr. 14, 2011.
Office action dated Dec. 5, 2012, U.S. Appl. No. 13/087,239, filed Apr. 14, 2011.
Office action dated Feb. 5, 2013, U.S. Appl. No. 13/092,724, filed Apr. 22, 2011.
Office action dated Jun. 10, 2013, U.S. Appl. No. 13/092,580, filed Apr. 22, 2011.
Office action dated Mar. 18, 2013, U.S. Appl. No. 13/040,2259, filed Mar. 7, 2011.
Office action dated Jun. 21, 2013, U.S. Appl. No. 13/092,460, filed Apr. 22, 2011.
Office action dated Jan. 28, 2013, U.S. Appl. No. 13/092,701, filed Apr. 22, 2011.
Office action dated Jul. 3, 2013, U.S. Appl. No. 13/092,701, filed Apr. 22, 2011.
Office action dated Jul. 18, 2013, U.S. Appl. No. 13/092,752, filed Apr. 22, 2011.
Office action dated Dec. 20, 2012, U.S. Appl. No. 12/950,974, filed Nov. 19, 2010.
Office action dated May 24, 2012, U.S. Appl. No. 12/950,974, filed Nov. 19, 2010.
Office action dated Sep. 5, 2013, U.S. Appl. No. 13/092,877, filed Apr. 22, 2011.
Office action dated Mar. 4, 2013, U.S. Appl. No. 13/092,877, filed Apr. 22, 2011.
Office action dated Jan. 4, 2013, U.S. Appl. No. 12/950,968, filed Nov. 19, 2010.
Office action dated Jun. 7, 2012, U.S. Appl. No. 12/950,968, filed Nov. 19, 2010.
Office action dated Sep. 19, 2012, U.S. Appl. No. 13/092,864, filed Apr. 22, 2011.
Office action dated May 31, 2013, U.S. Appl. No. 13/098,360, filed Apr. 29, 2011.
Office action dated Dec. 3, 2012, U.S. Appl. No. 13/030,806, filed Feb. 18, 2011.
Office action dated Apr. 22, 2014, U.S. Appl. No. 13/030,806, filed Feb. 18, 2011.
Office action dated Jun. 11, 2013, U.S. Appl. No. 13/030,806, filed Feb. 18, 2011.
Office action dated Apr. 25, 2013, U.S. Appl. No. 13/030,688, filed Feb. 18, 2011.
Office action dated Feb. 22, 2013, U.S. Appl. No. 13/044,301, filed Mar. 9, 2011.
Office action dated Jun. 11, 2013, U.S. Appl. No. 13/044,301, filed Mar. 9, 2011.
Office action dated Oct. 26, 2012, U.S. Appl. No. 13/050,102, filed Mar. 17, 2011.
Office action dated May 16, 2013, U.S. Appl. No. 13/050,102, filed Mar. 17, 2011.
Office action dated Aug. 4, 2014, U.S. Appl. No. 13/050,102, filed Mar. 17, 2011.
Office action dated Jan. 28, 2013, U.S. Appl. No. 13/148,526, filed Jul. 16, 2011.
Office action dated May 22, 2013, U.S. Appl. No. 13/148,526, filed Jul. 16, 2011.
Office action dated Jun. 19, 2013, U.S. Appl. No. 13/092,873, filed Apr. 22, 2011.
Office action dated Jul. 18, 2013, U.S. Appl. No. 13/365,808, filed Feb. 3, 2012.
Office action dated Jun. 13, 2013, U.S. Appl. No. 13/312,903, filed Dec. 6, 2011.
Lapuh, Roger et al., ‘Split Multi-link Trunking (SMLT) draft-lapuh-network-smlt-08’, Jan. 2009.
Office Action for U.S. Appl. No. 13/030,688, filed Feb. 18, 2011, dated Jul. 17, 2014.
Office Action for U.S. Appl. No. 13/044,326, filed Mar. 9, 2011, dated Jul. 7, 2014.
Office Action for U.S. Appl. No. 13/092,752, filed Apr. 22, 2011, dated Apr. 9, 2014.
Office Action for U.S. Appl. No. 13/092,873, filed Apr. 22, 2011, dated Jul. 25, 2014.
Office Action for U.S. Appl. No. 13/092,877, filed Apr. 22, 2011, dated Jun. 20, 2014.
Office Action for U.S. Appl. No. 13/312,903, filed Dec. 6, 2011, dated Aug. 7, 2014.
Office Action for U.S. Appl. No. 13/351,513, filed Jan. 17, 2012, dated Jul. 24, 2014.
Office Action for U.S. Appl. No. 13/425,238, filed Mar. 20, 2012, dated Mar. 6, 2014.
Office Action for U.S. Appl. No. 13/556,061, filed Jul. 23, 2012, dated Jun. 6, 2014.
Office Action for U.S. Appl. No. 13/742,207 dated Jul. 24, 2014, filed Jan. 15, 2013.
Office Action for U.S. Appl. No. 13/950,974, filed Nov. 19, 2010, from Haile, Awet A., dated Dec. 2, 2012.
Office Action for U.S. Appl. No. 13/087,239, filed Apr. 14, 2011, dated Dec. 5, 2012.
Office Action for U.S. Appl. No. 13/351,513, filed Jan. 17, 2012, dated Feb. 28, 2014.
Perlman R: ‘Challenges and opportunities in the design of TRILL: a routed layer 2 technology’, 2009 IEEE Globecom Workshops, Honolulu, HI, USA, Piscataway, NJ, USA, Nov. 30, 2009 (Nov. 30, 2009), pp. 1-6, XP002649647, DOI: 10.1109/GLOBECOM.2009.5360776 ISBN: 1-4244-5626-0 [retrieved on Jul. 19, 2011].
TRILL Working Group Internet-Draft Intended status: Proposed Standard RBridges: Base Protocol Specificaiton Mar. 3, 2010.
Office action dated Aug. 14, 2014, U.S. Appl. No. 13/092,460, filed Apr. 22, 2011.
Office action dated Jul. 7, 2014, for U.S. Appl. No. 13/044,326, filed Mar. 9, 2011.
Office Action dated Dec. 19, 2014, for U.S. Appl. No. 13/044,326, filed Mar. 9, 2011.
Office Action for U.S. Appl. No. 13/092,873, filed Apr. 22, 2011, dated Nov. 7, 2014.
Office Action for U.S. Appl. No. 13/092,877, filed Apr. 22, 2011, dated Nov. 10, 2014.
Office Action for U.S. Appl. No. 13/157,942, filed Jun. 10, 2011.
Mckeown, Nick et al. “OpenFlow: Enabling Innovation in Campus Networks”, Mar. 14, 2008, www.openflow.org/documents/openflow-wp-latest.pdf.
Office Action for U.S. Appl. No. 13/044,301, dated Mar. 9, 2011.
Office Action for U.S. Appl. No. 13/184,526, filed Jul. 16, 2011, dated Jan. 5, 2015.
Office Action for U.S. Appl. No. 13/598,204, filed Aug. 29, 2012, dated Jan. 5, 2015.
Office Action for U.S. Appl. No. 13/669,357, filed Nov. 5, 2012, dated Jan. 30, 2015.
Office Action for U.S. Appl. No. 13/851,026, filed Mar. 26, 2013, dated Jan. 30, 2015.
Office Action for U.S. Appl. No. 13/786,328, filed Mar. 5, 2013, dated Mar. 13, 2015.
Office Action for U.S. Appl. No. 13/092,460, filed Apr. 22, 2011, dated Mar. 13, 2015.
Office Action for U.S. Appl. No. 13/425,238, dated Mar. 12, 2015.
Office Action for U.S. Appl. No. 13/092,752, filed Apr. 22, 2011, dated Feb. 27, 2015.
Office Action for U.S. Appl. No. 13/042,259, filed Mar. 7, 2011, dated Feb. 23, 2015.
Office Action for U.S. Appl. No. 13/044,301, filed Mar. 9, 2011, dated Jan. 29, 2015.
Office Action for U.S. Appl. No. 13/050,102, filed Mar. 17, 2011, dated Jan. 26, 2015.
Rosen, E. et al., “BGP/MPLS VPNs”, Mar. 1999.
Office Action for U.S. Appl. No. 14/577,785, filed Dec. 19, 2014, dated Apr. 13, 2015.
Office Action for U.S. Appl. No. 13/425,238, filed Mar. 20, 2012, dated Mar. 12, 2015.
Abawajy J. “An Approach to Support a Single Service Provider Address Image for Wide Area Networks Environment” Centre for Parallel and Distributed Computing, School of Computer Science Carleton University, Ottawa, Ontario, K1S 5B6, Canada.
Office Action dated Jan. 31, 2017, U.S. Appl. No. 13/184,526, filed Jul. 16, 2011.
Office Action dated Jan. 27, 2017, U.S. Appl. No. 14/216,292, filed Mar. 17, 2014.
Office Action dated Jan. 26, 2017, U.S. Appl. No. 13/786,328, filed Mar. 5, 2013.
Office Action dated Dec. 2, 2016, U.S. Appl No. 14/512,268, filed Oct. 10, 2014.
Office Action dated Dec. 1, 2016, U.S. Appl. No. 13/899,849, filed May 22, 2013.
Office Action dated Dec. 1, 2016, U.S. Appl. No. 13/656,438, filed Oct. 19, 2012.
Office Action dated Nov. 30, 2016, U.S. Appl. No. 13/598,204, filed Aug. 29, 2012.
Office Action dated Nov. 21, 2016, U.S. Appl. No. 13/669,357, filed Nov. 5, 2012.
Office Action dated Feb. 8, 2017, U.S. Appl. No. 14/473,941, filed Aug. 29, 2014.
Office Action dated Feb. 8, 2017, U.S. Appl. No. 14/822,380, filed Aug. 10, 2015.
Office Action dated 06/18/215, U.S. Appl. No. 13/098,490, filed May 2, 2011.
Office Action dated Jul. 31, 2015, U.S. Appl. No. 13/598,204, filed Aug. 29, 2014.
Office Action dated Jul. 31, 2015, U.S. Appl. No. 14/473,941, filed Aug. 29, 2014.
Office Action dated Jul. 31, 2015, U.S. Appl. No. 14/488,173, filed Sep. 16, 2014.
Office Action dated Aug. 21, 2015, U.S. Appl. No. 13/776,217, filed Feb. 25, 2013.
Office Action dated Aug. 19, 2015, U.S. Appl. No. 14/156,374, filed Jan. 15, 2014.
Office Action dated Sep. 2, 2015, U.S. Appl. No. 14/151,693, filed Jan. 9, 2014.
Office Action dated Sep. 17, 2015, U.S. Appl. No. 14/577,785, filed Dec. 19, 2014.
Office Action dated Sep. 22, 2015 U.S. Appl. No. 13/656,438, filed Oct. 19, 2012.
Office Action dated Nov. 5, 2015, U.S. Appl. No. 14/178,042, filed Feb. 11, 2014.
Office Action dated Oct. 19, 2015, U.S. Appl. No. 14/215,996, filed Mar. 17, 2014.
Office Action dated Sep. 18, 2015, U.S. Appl. No. 13/345,566, filed Jan. 6, 2012.
Office Action dated Jun. 13, 2018, U.S. Appl. No. 13/786,328, filed Mar. 5, 2013.
IEEE et al., “Amendment to Carrier Sense Multiple Access with Collision Detection (CSMA/CD) Access Method and Physical Layer Specifications—Aggregation of Multiple Link Segments”, Mar. 30, 2000, IEEE Computer Society, IEEE Std 802.3ad-2000, pp. 116-117.
Office Action dated Jul. 13, 2018, U.S. Appl. No. 15/402,924, filed Jul. 13, 2018.
Office Action dated Jul. 24, 2018, U.S. Appl. No. 14/799,371, filed Jul. 24, 2018.
Office Action for U.S. Appl. No. 14/216,292, dated Oct. 6, 2017.
Office Action dated Oct. 25, 2017, U.S. Appl. No. 14/867,865, filed Sep. 28, 2015.
Office action dated Oct. 26, 2017, U.S. Appl. No. 14/817,097, filed Aug. 3, 2015.
Office Action dated Mar. 20, 2018, U.S. Appl. No. 14/867,865, filed Sep. 28, 2015.
Office Action for U.S. Appl. No. 14/817,097 dated May 4, 2017.
Office Action for U.S. Appl. No. 14/872,966, dated Apr. 20, 2017.
Office Action for U.S. Appl. No. 14/680,915, dated May 3, 2017.
Office Action for U.S. Appl. No. 14/792,166, dated Apr. 26, 2017.
Office Action for U.S. Appl. No. 14/510, 913, dated Jun. 30, 2017.
Office Action for U.S. Appl. No. 15/005,946, dated Jul. 14, 2017.
Office Action for U.S. Appl. No. 13/092,873, dated Jul. 19, 2017.
Office Action for U.S. Appl. No. 15/047,539, dated Aug. 7, 2017.
Office Action for U.S. Appl. No. 14/830,035, dated Aug. 28, 2017.
Office Action for U.S. Appl. No. 13/098,490, dated Aug. 24, 2017.
Office Action for U.S. Appl. No. 13/786,328, dated Aug. 21, 2017.
Office Action for U.S. Appl. No. 14/660,803, dated May 17, 2017.
Office Action for U.S. Appl. No. 14/488,173, dated May 12, 2017.
Office Action for U.S. Appl. No. 13/288,822, dated May 26, 2017.
Office Action for U.S. Appl. No. 14/329,447, dated Jun. 8, 2017.
“Network based IP VPN Architecture using Virtual Routers” Paul Knight et al.
Yang Yu et al “A Framework of using OpenFlow to handle transient link failure”, TMEE, 2011 International Conference on, IEEE, Dec. 16, 2011.
Office Action for U.S. Appl. No. 15/227,789, dated Feb. 27, 2017.
Office Action for U.S. Appl. No. 14/822,380, dated Feb. 8, 2017.
Office Action for U.S. Appl. No. 14/704,660, dated Feb. 27, 2017.
Office Action for U.S. Appl. No. 14/510,913, dated Mar. 3, 2017.
Office Action for U.S. Appl. No. 14/473,941, dated Feb. 8, 2017.
Office Action for U.S. Appl. No. 14/329,447, dated Feb. 10, 2017.
Office Action for U.S. Appl. No. 14/662,095, dated Mar. 24, 2017.
Office Action for U.S. Appl. No. 15/005,967, dated Mar. 31, 2017.
Office Action for U.S. Appl. No. 15/215,377, dated Apr. 7, 2017.
Office Action for U.S. Appl. No. 13/098,490, dated Apr. 6, 2017.
Office Action for U.S. Appl. No., 14/662,092, dated Mar. 29. 2017.
Related Publications (1)
Number Date Country
20150271104 A1 Sep 2015 US
Provisional Applications (1)
Number Date Country
61968141 Mar 2014 US