Patent Application
20140185607
The present application claims priority from Japanese application JP 2012-286631 filed on, Dec. 28,2012, the content of which is hereby incorporated by reference into this application.
The present invention relates to a transmission system, and in particular to a system for coupling to a communication network to form communication paths by autonomous distributed control in order to send and receive data by way of a transmission network utilizing centralized control to establish communication paths.
Telecommunications carriers often provide services to the user over networks utilizing devices forming communication paths by autonomous distributed control. For example, IP/MPLS (Internet Protocol/Multi-Protocol Label Switching) routers are communication devices based on MPLS technology that are utilized in particular in core networks of telecommunications carriers. Each IP/MPLS node includes both a C plane (Control plane) as a function for controlling network paths, and a D plane (Data plane) as a function for transferring data, and is featured in being capable of dynamically setting up logical paths (referred to simply as “paths” in these specification) indicating communication paths under autonomous distributed control, conforming to paths determined by IP (Internet Protocol) routing. The IP/MPLS router contains a function to couple the stored IP service with the established paths. The IP/MPLS router can in this way flexibly provide End-to-End connectivity according to the stored IP service.
On the other hand, increasing demands for higher network efficiency and reliability have in some case led to transmission networks utilizing transmission devices that form communication paths by centralized control. For example, MPLS-TP (Multi-Protocol Label Switching-Transport Profile) is a transmission technology based on MPLS technology the same as IP/MPLS. Each node MPLS-TP node includes a D plane for user data flow, but does not include a C plane for controlling paths for user data flow. Namely a feature of this transmission technology is that the C plane and D plane are isolated and that paths are established by centralized control. Therefore, compared to IP/MPLS, MPLS-TP is capable of configuring a network that transfers packets with high efficiency by taking into account overall network traffic characteristics. Moreover, MPLS-TP is fully equipped with high reliability functions such as OAM (Operation Administration Maintenance) function or APS (Automatic Protection Switching) functions serving as a network infrastructure for storing network services. Since these types of advantages are available, the use of both networks in interworking technology or namely using MPLS-TP networks to couple IP/MPLS networks together in order to achieve high speed in IP services provided by IP/MPLS is being discussed by standards organizations, etc.
Achieving End-to-End connectivity interworking systems using MPLS-TP networks and IP/MPLS networks requires mapping to link both paths together. R. Martinotti, et al., “Interworking between MPLS-TP and IP/MPLS”, draft-martinotti-mpls-tp-interworking-02, Internet-Draft, IETF, 2011 describes two types of path mapping systems respectively called the overlay model and peer model.
In the overlay model, the MPLS-TP network transparently couples between the IP/MPLS networks. In the edge device for the MPLS-TP network, the paths for the MPLS-TP network are mapped relative to VLAN (Virtual Location Area Network) tags attached to the packets in the IP/MPLS network or the physical IF (Interface) of the coupled edge devices. The plural coupled IP/MPLS network paths are therefore mapped as a single MPLS-TP path.
Unlike the overlay model on the other hand, in the peer model the IP/MPLS network and the MPLS/TP networks are coupled on the same layer (MPLS path). Within the MPLS-TP network, the MPLS-TP path, and the IP/MPLS paths formed as needed are mapped in a 1-to-1 relationship.
In the overlay model in R. Martinotti, et al., “Interworking between MPLS-TP and IP/MPLS”, draft-martinotti-mpls-tp-interworking-02, Internet-Draft, IETF, 2011, each edge router in the IP/MPLS network recognizes that adjacent routers are mutually coupled without being aware of the MPLS-TP network. Therefore, when the system scale expands, there is a sudden increase in routing information that must be managed in each edge router so implementing large-scales systems with this overlay model is difficult.
In the peer model described in R. Martinotti, et al., “Interworking between MPLS-TP and IP/MPLS”, draft-martinotti-mpls-tp-interworking-02, Internet-Draft, IETF, 2011, the IP/MPLS paths and MPLS-TP paths are mapped in a 1-to-1 relationship so that even if a paths were routed between edge devices in the same MPLS-TP network, each individual MPLS-TP path having a regular bandwidth had to be mapped relative to all of the IP/MPLS paths. Due to this necessity, the amount of unused bandwidth increased, consequently resulting in a drop in the MPLS-TP network utilization rate.
Whereupon in view of the aforementioned problems, the present invention has the object of providing a mapping model ideal for mapping paths in large-scale systems and ideal for mapping paths achieving a transmission network having a high bandwidth utilization rate; in interworking systems utilizing a transmission network (such as MPLS/TP networks) of paths organized by centralized control; to connect between communication networks (such as IP/MPLS networks) having communication paths structured by autonomous distributed control.
To achieve the aforementioned objects, one aspect of the present invention provides a communication system comprised of a plurality of transmission devices configuring a transmission network conforming to a first communication protocol and coupled to a communication device configuring a communication network conforming to a second protocol, and a management server coupled to the transmission devices. The management server in the relevant communication system is configured from a first path information relating to a first communication path serving as a communication path between each of the preset transmission devices; and a path setter unit to select a corresponding first communication path based on the first path information; according to transmission devices comprising the transmission network passing along a second transmission path that is established by way of a transmission network serving as each of the start points and end points in either or both of two communication devices; and a transmit unit to transmit information relating to a first communication path selected by the path setter unit to the transmission device; and the transmission device in the communication system includes a send-receive unit to receive information relating to the first communication path, and a transfer processing unit to execute transfer processing of information received from the communication device based on the received information relating to the first communication path.
More preferably, the path setter unit maps (links) the common first communication path to the plurality of second communication paths passing along the common transmission devices, based on the first path information.
Even more preferably, the send-receive unit includes a transmit unit to transfer control packets for setting up a communication path protocol in order to establish a second communication path that the transmission device received from the communication device; and the path setter unit decides based on the control packet if the second communication path passes along the transmission device, and links the first communication path to the second communication path based on the decision results and the first path information.
Another aspect of the present invention provides a communication path establishing method that sets up a communication path between communication devices configuring the protocol, by way of a transmission network comprised of plural transmission devices and in conformance to a first communication protocol. The communication path establishing method sets a first communication path serving as the communication path between each of edge devices coupled to communication devices among the transmission device, and selects a first communication path according to edge devices that a second transmission path passes through, corresponding to the second transmission path established by two communication devices serving as each of the start points and end points by way of a transmission network, and sets up the second communication path by way of the selected first communication path.
More preferably, the communication path establishing method links the common first communication path, to the plurality of second communication paths passing along the common edge devices.
Further, another aspect of the present invention provides a management server coupled to plural transmission devices and configuring a transmission network conforming to a first communication protocol, and coupled to communication devices configuring a communication network conforming to a second communication protocol. The management server is comprised of first path information relating to a first communication path serving as the communication path between each of the preset transmission devices; and a path setter unit to select a first communication path corresponding to a second transmission path established by two communication devices serving as each of the start points and end points by way of a transmission network, based on the first path information, according to a transmission device configuring the transmission networks through which a second transmission path passes; and a transmit unit to transmit information relating to a first communication path selected by the path setter unit to the transmission device.
The present invention is capable of achieving a system having an excellent utilization rate and scalability in a transmission system for interworking with communication networks. The present invention is also simultaneously capable of achieving a transmission system that provides a communication service ensuring excellent guaranteed bandwidth.
The embodiments of the present invention are described by an example for the case where the MPLS-TP network is utilized as the transmission network, however other networks utilizing a protocol for establishing communication paths by centralized control may also be used and the same effect obtained even if another network is utilized. An example of the case where utilizing an IP/MPLS network as the communication network is described but other networks utilizing a protocol for establishing communication paths by centralized control may also be used and the same effect obtained even if using another network is utilized.
An example of the system structure for applying the present invention is shown in
The embodiment of the present invention is hereafter described in detail.
The device control unit 340 is coupled to the centralized control server 100, and includes a function to set the received setting information into the switching unit 350, the IP/MPLS network IF 310, and the MPLS-TP network IF 320. The device control unit 340 also includes a function to send and receive the control packets for the routing protocols for setting the paths for the IP layer of the IP/MPLS and for the signaling protocol to establish the MPLS paths, to and from the centralized control server 100.
The switching unit 350 specifies a transfer destination IF from the MPLS label and the transfer source IF of the packets, and transfers the packet to the appropriate IP/MPLS network IF 310 or MPLS-TP network IF 320.
The IP/MPLS network IF 310 is an IF coupled to the IP/MPLS router 200 and is comprised of a receive circuit 311, a L2 receive processor unit 312, a packet analysis-assignment unit 313, a SW transmit circuit 314, a SW receive circuit 315, a packet cluster unit 316, a L2 transmit processor unit 317, a transmit circuit 318, and an IF control unit 319.
The IF control unit 319 couples to the device control unit 340 and includes a function to set the setting information notified from the device control unit 340 into each structural unit of the IP/MPLS network IF 310; and a function to load (read-out) information set in each structural unit and notify the device control unit 340. The IF control unit 319 also includes a function that sends and receives control packets for IP/MPLS routing protocols and for signaling protocols to and from the device control unit 340.
The receive circuit 311 is coupled to the IP/MPLS router 200 and receives packets from the IP/MPLS router 200.
The L2 receive processor unit 312 is the end point for data link layer protocols of the Open Systems Interconnection (OSI) reference model coupling between the IP/MPLS router 200 and the MPLS-TP device 300. When the data link layer protocol is for example the Ethernet, the L2 receive processor unit 312 internally executes the Ethernet frame end point processing. Moreover, the L2 receive processor unit 312 learns the transmit source MAC address of the received Ethernet frame and shares the information with the L2 transmit processor unit 317.
The packet analysis-assignment unit 313 analyzes the packet received from the L2 receive processor unit 312 and when the received packet is a data packet, transfers the received data to the SW transmit circuit 314. When the received packet is a control packet for the IP/MPLS routing protocol and signaling protocol, the packet analysis-assignment unit 313 transfers the relevant packet by way of the IF control unit 319 to the centralized control server 100.
The SW transmit circuit 314 transfers the packet received from the packet analysis-assignment unit 313 to the switching unit 350.
The SW receive circuit 315 receives the packet from the switching unit 350 and transfers the packet to the packet cluster unit 316.
After the packet cluster unit 316 receives the data packet from the SW receive circuit 315, the packet cluster unit 316 analyzes the received packet and transfers the packet to the L2 transmit processor unit 317. The packet cluster unit 316 transfers the (IP/MPLS) routing protocol control packet and signaling protocol control packet that were received from the centralized control server 100 by way of the IF control unit 319 to the L2 transmit processor unit 317.
When the packet is received, the L2 transmit processor unit 317 generates a MAC header from the MAC address information jointly shared with the L2 receive processor unit 312, attaches the MAC header to the packet, and transfers the packet to the transmit circuit 318.
The transmit circuit 318 sends the packet received from the L2 transmit processor unit 317 to the IP/MPLS router 200.
The MPLS-TP network IF 320 is an interface coupling to other MPLS-TP devices 300 configuring the MPLS-TP network 30. The MPLS-TP network IF 320 is comprised of a receive circuit 321, a MPLS header converter unit 322, an IP/MPLS network label converter unit 323, a SW transmit circuit 324, a SW receive circuit 325, a MPLS header insertion unit 326, a transmit circuit 327, a MPLS-TP label database 328, an IP/MPLS label database 329, an IF control unit 330, a scheduler 331, and a bandwidth control processor unit 332.
The IF control unit 330 couples to the device control unit 340 and includes a function to set the setting information reported from the device control unit 340 in each structural unit of the MPLS-TP networks IF 320, and a function to load (read) the information set in each structural unit and notify the device control unit 340.
The receive circuit 321 includes a function to receive packets from the other MPLS-TP devices 300.
The MPLS header converter unit 322 searches the MPLS-TP label information table 550 of the MPLS-TP label database 328. The MPLS header converter unit 322 includes a function to convert the packet label.
The IP/MPLS network label converter unit 323 contains a function to search the IP/MPLS label information table 560 in the IP/MPLS label database 329 and convert the IP/MPLS label of the packet.
The scheduler 331 performs output arbitration of the packet received from the MPLS header converter unit 322 and the IP/MPLS network label converter unit 323 and transfers it to the SW transmit circuit 324.
The SW transmit circuit 324 transfers the packet received from the scheduler 331 to the switching unit 350.
The SW receive circuit 325 receives the packet from the switching unit 350, and transfers it to the MPLS header insertion unit 326.
The MPLS header insertion unit 326 searches the IP/MPLS input label 551 of the MPLS-TP label information table 550 for a match with the label attached to the beginning of the received packet. When an applicable entry is found, the MPLS header insertion unit 326 newly inserts a MPLS header at the beginning, records the MPLS-TP output label 553 corresponding to that header and transfers it to the transmit circuit 327. When there is no applicable entry, the received packet is transferred unchanged to the transmit circuit 327.
The bandwidth control processor unit 332 contains a function for bandwidth control processing based on the guaranteed bandwidth 555 for each path in the MPLS-TP network registered in the MPLS-TP label information table 550.
The transmit circuit 327 contains a function to send the packet received from the MPLS header insertion unit 326 to the MPLS-TP network 30 side.
The centralized control server 100 of the present invention is described next while referring to
The device setter unit 110 is coupled to the MPLS-TP device 300 and contains a function to transfer device setting information from the setting processor unit 120, and control packets for routing protocols and signaling protocols. The device setter unit 110 also contains a function to receive control packets for IP/MPLS signaling protocols and routing protocols sent from the MPLS-TP device 300 and transfer them to the setting processor unit 120.
The administrator setting unit 130 contains a function to receive messages from external sources when the administrator makes static settings in the centralized control server 100 and send them to the setting processor unit 120.
The setting processor unit 120 contains a function to send device setting information, and control packets for routing protocols and signaling protocols to the device setter unit 110. The setting processor unit 120 also contains a function to judge the control packet received from the device setting unit 110, and select and send a suitable control packet from the routing processor unit 140 and the path setter unit 141.
The routing processor unit 140 contains a function to process the IP/MPLS routing protocol control packet that was received, and a function to generate and send a control packet for the IP/MPLS router 200. An OSPF (Open Shortest Path First) or RIP (Routing Information Protocol) and others may for example be utilized as the routing protocol. The processing of this routing protocol forms a routing information table 500 in the routing database 150.
The path setter unit 141 includes a function to process the signaling protocol received by way of the MPLS-TP device from the IP/MPLS network 20. The LDP (Label Distribution Protocol) and the RSVP-TE (Resource Reservation Protocol-Traffic Engineering), etc. may for example be utilized as the signaling protocol. The path setter unit 141 maps the path within the MPLS-TP network 30 relative to the path in the IP/MPLS network 20 established by the signaling protocol by referring to the routing database 150, the link information database 151, and the path information database 152, and registers the path in the path information database 152. The path setter unit 141 also includes a function to send messages to operate the MPLS-TP label database 328 and IP/MPLS label database 329 of the applicable MPLS-TP device 300 according to the information in the path information database 152.
The link information database 151 includes a link information table 510.
The path information database 152 contains the MPLS-TP network path information table 520 and the mapping information table 530.
The operation of the first embodiment is described while referring to the examples. In this example of the operation as shown in
First of all, before establishing the IP/MPLS paths, the routing processor unit 140 of the centralized control server 100 sends and receives control packets for the routing protocol with the IP/MPLS router 200 by way of the MPLS-TP device 300 on the network edge and processes the control packets to generate a routing information table 500 and retains the routing information table 500 made by the routing database 150. The administrator operates the path setter unit 141 by way of the administrator setting unit 130, and registers the default paths to connect between all of the edge devices assumed to pass along when setting up paths for the IP/MPLS network, into the MPLS-TP network path information table 520 of the path information database 152. There is no priority level setting for the LDP of the signaling protocol utilized in the present embodiment so the default path types are all stationary (fixed) defaults. The path setter unit 141 registers label information regarding the default paths registered in the path information database 152 into the MPLS-TP label information table 550 of the MPLS-TP device 300 as the edge device serving as the default path start point and end point. Here, prior to the start of mapping, the entries in the MPLS-TP label information table 550 of the edge device serving as the start point of the default path is in a state where there is no IP/MPLS input label 551 matching the relevant default path; and a pop instruction is registered in the MPLS-TP output label 553 corresponding to the MPLS-TP input label 552.
An example of the operation for mapping of paths linked to the signaling protocol is hereafter described while referring to the sequence in
Hereafter, when the MPLS-TP device 300-1 receives the data packet 602 attached with a label 10 at the beginning from the IP/MPLS router 200-2, after a switch is made to the appropriate MPLS-TP network IF320 by the switching unit 350 based on the label, the MPLS header insertion unit 326 searches the MPLS-TP label information table 550 based on the label of the packet. The 10 here is applicable to the IP/MPLS input label 551 and so the MPLS header is added to the packet, the matching MPLS-TP output label 553 is attached, and transferred to the MPLS-TP device 300-2 by the default path (S106). The MPLS-TP device 300-2 serving as the output device searches the MPLS-TP label information table 550 based on the label of the packet in the MPLS-TP header converter unit 322 of the MPLS-TP network IF320. A pop instruction corresponding to the attached MPLS-TP input label 552 is then found so the beginning MPLS-TP header is deleted and the packet is transferred to the IP/MPLS network label converter unit 323. The IP/MPLS network label converter unit 323 searches the IP/MPLS label information table 560, and converts (swaps) the IP/MPLS input label 561 attached to the packet, for the corresponding IP/MPLS output label 562 (20), and transfers the packet from the appropriate IP/MPLS network IF to the IP/MPLS router 200-3.
The operations S108, S109, S110 when the IP/MPLS router 200-7 sends a label allocation message 604 to the MPLS-TP device 300-2, and the centralized control server 100 sends a label allocation message 605 by way of the MPLS-TP device 300-1 to the IP/MLS router 200-6, are the same as the operations S101, S102, S103. The operation in S111 is the same as in S104 but in terms of results, the entries for the MPLS-TP network path ID532, input device ID533, input IF ID534, output device ID535, and output device IF ID536 made in S104 as common entries are made in S111; and the input label (30 in the present embodiment) assigned to the IP/MPLS router 200-6 in S111 is additionally registered in the entry of the IP/MPLS input label 551 registered in S104 in the MPLS-TP label information table 550 of the MPLS-TP device 300-1. In this way, when the MPLS-TP device 300-1 has received a data packet 605 with a label 30 attached, by the adding of a MPLS header, a packet is transferred to the MPLS-TP device 300-2 on a common default path with S106 (S113), the MPLS header is deleted and the labels swapped, and the packet sent to the IP/MPLS router 200-3.
The above operation allows mapping of common default paths and transferring the packets when the edge MPLS-TP devices 300 along the route are the same on the paths of different IP/MPLS networks so that the utilization rate of the MPLS-TP network can be improved. In the case of LDP, the operation after establishing the IP/MPLS input label and output label for an optional IP/MPLS network path is the same even in the case where utilizing an assignment method different from the present embodiment, and the same effect can be obtained.
A feature of the second embodiment is a decision to set mapping to default path or mapping to separate paths by order of priority when path mapping by utilizing RSVP-TE to establish guaranteed bandwidth paths, as a signaling protocol. In RSVP-TE, establishing of paths starts by transferring path messages including the address of the IP/MPLS router 200 farthest downstream, the requested bandwidth, the order of priority (e.g. setup priority) and so on, from the IP/MPLS router on the highest upstream path in the IP/MPLS network to the destination IP/MPLS router. The path with the procured bandwidth is next established by transferring a resource scheduling message including requested bandwidth, and IP/MPLS label attached when sending to the next hop, from the farthest downstream IP/MPLS router that received the path message, to the IP/MPLS router that is farthest upstream. The route of the IP/MPLS network path established by RSVP-TE can be clearly specified within the path message, and non-specified zones are specified in accordance with the routing table.
An example of the operation is described next. The system structure utilized is identical to that of the first embodiment and is described in the example of
An operation example relating to consecutive path mapping with RSVP-TE is described using the sequence in
When the MPLS-TP device 300-1 receives the path message 616 in S216-S220, the message is transferred to the centralized control server 100; and the path setter unit 141 acquires and retains the priority, the requested bandwidth, the destination address, and the session ID of the paths in the IP/MPLS network established by way of the path message 610, and the device ID and input IF ID and so on of the MPLS-TP device 300-1 that received the path message 616 serving as the input device for the established paths in the IP/MPLS network, the same as in the previous operation. Here, the path setter unit 141 is assumed to have decided the high order of priority. In this case, the path setter unit 141 decides that mapping can be performed on separate paths that satisfy the requested bandwidth among edge devices for passing through. For example, one or more separate paths can be statically set between the edge devices, the path setter unit 141 searches the MPLS-TP network path information table 520 and the mapping information table 530, and decides whether or not there are separate paths satisfying the requested bandwidth among the relevant edge devices. The path setter unit 141 can also search the link information table 510, set a new separate path satisfying the requested bandwidth, and decide if mapping is possible. If decided here that mapping is impossible then the following operation is not performed. If the path setter unit 141 decides that mapping is possible, the path message is sent to the IP/MPLS router 200-7 at the next hop the same as the previous operation. In S221-S225, the path setter unit 141 of the centralized control server 100 processes the resource scheduling message the same as in S205-S209, and registers the new entries in the mapping information table 530. The separate paths are next mapped rather than the default path the same operation as in S210. Here, if new separate paths were established, then the surplus bandwidth is change according to the guaranteed bandwidth for the newly established separate paths in the link information table 510. In S211-S215, the data packet 620 is transferred within the MPLS-TP network by utilizing the separate paths by the same operation as in S105-S107. The above operation handles low priority paths by mapping the common default path, and handles high priority paths by mapping separate paths in a one-to-one relation so that paths requiring the exact guaranteed bandwidth can be acquired while improving the utilization efficiency of the MPLS-TP network.
| Number | Date | Country | Kind |
|---|---|---|---|
| 2012-286631 | Dec 2012 | JP | national |
