Graceful scaling in software driven networks

Abstract

Provided are methods and systems for graceful scaling of data networks. In one example, an indication of removal of a node from a plurality of nodes of the data network is received. A service policy is generated to reassign service requests associated with the node to another node in the plurality of nodes. The service policy is then sent to each of the plurality of nodes of the data network. To scale out a data network, an indication of presence of a further node in the data network is received, and a further node service policy is generated and sent to each of the plurality of nodes of the data network and to the further node. Additional actions can be taken in order to prevent interruption of an existing heavy-duty connection while scaling the data network.

Description

TECHNICAL FIELD

This disclosure relates generally to data processing and, more specifically, to methods and system for scaling data networks.

BACKGROUND

The approaches described in this section could be pursued but are not necessarily approaches that have previously been conceived or pursued. Therefore, unless otherwise indicated, it should not be assumed that any of the approaches described in this section qualify as prior art merely by virtue of their inclusion in this section.

In a typical load balancing scenario, a service hosted by a group of servers is front-ended by a load balancer (LB) (also referred to herein as a LB device) which represents this service to clients as a virtual service. Clients needing the service can address their packets to the virtual service using a virtual Internet Protocol (IP) address and a virtual port. The LB will inspect incoming packets and, based on the policies/algorithms, will choose a particular server from the group of servers, modify the packet if needed, and forward the packet towards the server. On the way back from the server (optional), the LB will get the packet, modify the packet if needed, and forward the packet back towards the client.

The traditional approach for load balancing of network of servers has several drawbacks. For example, the network request load may stay lower than a maximum capacity of a LB device for a long time, which could lead to wasted resources. In another situation, a network request might exceed the maximum capacity a single LB device can handle. Generally speaking, the limitations of traditional load balancing of networks are due to there only being one or several static devices responsible for deciding how and where to send packets, which does not allow for dynamic changes in network configuration when one needs to scale down or scale up the resources.

Therefore, more efficient methods and systems for scaling data networks may be needed.

SUMMARY

This summary is provided to introduce a selection of concepts in a simplified form that are further described in the Detailed Description below. This summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used as an aid in determining the scope of the claimed subject matter.

The present disclosure is related to approaches for scaling in a data network (for example, a software driven network (SDN)). Specifically, a method for scaling in a data network may comprise receiving an indication of removing a first node from a plurality of nodes of the data network, generating a service policy, and sending the service policy to each of the plurality of nodes of the data network. The service policy reassigns service requests associated with the first node to a second node from the plurality of nodes of the data network. The first node is removed from the data network and the second node remains in the data network. The service policy can include a classification traffic map, wherein classes of service requests are assigned to nodes from the plurality of nodes of the data network.

In some embodiments, the method for scaling in a data network may include additional steps preventing interruption of heavy-duty connections served by the first node. For instance, before sending the service policy to each of the plurality of nodes of the data network, an indication of a presence of old connections associated with the first node can be received, and a redirection node can be created in a data network. A redirection policy can then be generated. The redirection policy contains an indication that service requests associated with the old connections should be sent from the second node via the redirection node to the first node. The redirection policy is sent to the second node and the redirection node. After sending the service policy to each of the plurality of nodes of the data network, the method for scaling in a data network can include waiting for an indication of finishing the old connection. Upon receiving the indication, the redirection policy is removed from the second node and the redirection node is removed from the data network.

In some embodiments, a method for scaling out a data network may comprise receiving an indication of a presence of a further node in the data network, generating a further node service policy, and sending the further node service policy to each of a plurality of nodes of the data network and to the further node. The further node service policy reassigns one or more of the service requests associated with any of the plurality of nodes of the data network to the further node.

In some embodiments, the method for scaling out a data network includes additional steps preventing interruption of heavy-duty connections served by one or more old nodes in plurality of the nodes of the data network. For example, before sending the further node service policy to each of the plurality of nodes of the data network, the method may include receiving an indication of a presence of old connections associated with an old node from the plurality of nodes of the data network. A redirection node can be created in the data network, and a redirection policy can be generated. The redirection policy indicates that service requests associated with the old connections should be sent from the further node via the redirection node to the old node. The redirection policy is sent to the redirection node and to the further node. After sending the further node service policy to each of the plurality of nodes of the data network and to the further node, the method may include waiting for an indication of finishing the old connections. Upon receiving the indication of finishing the old connections, the redirection policy is removed from the further node and the redirection node is removed from the data network.

According to another example embodiment, there is provided a system for scaling a data network. The system includes a cluster master and a plurality of nodes of a data network. The cluster master is configured to retrieve and analyze network data associated with the data network and service node data associated with one or more service nodes. The data network may be scaled in. For example, the cluster master can receive an indication of removing a first node from the plurality of nodes of the data network, generate a service policy, and send the service policy to each of the plurality of nodes of the data network. The service policy reassigns service requests associated with the first node to a second node from the plurality of nodes of the data network.

In other embodiments, the data network may be scaled out. For example, the cluster master can receive an indication of presence of a further node in the data network, generate a further node service policy, and send the further node service policy to each of the plurality of nodes of the data network and to the further node. The further node service policy reassigns one or more of the service requests associated with any of existing nodes of the plurality of nodes of the data network to the further node. In certain embodiments, the cluster master can be configured to carry out additional steps preventing interruption of heavy-duty connections served by the first node when the data network is scaled in or by one of the existing nodes when the data network scaled out.

In further example embodiments of the present disclosure, the steps of a method for scaling a data networks are stored on a machine-readable medium comprising instructions, which when implemented by one or more processors perform the recited steps. In yet further example embodiments, hardware systems or devices can be adapted to perform the recited steps. Other features, examples, and embodiments are described below.

BRIEF DESCRIPTION OF THE DRAWINGS

Embodiments are illustrated by way of example, and not by limitation, in the figures of the accompanying drawings, in which like references indicate similar elements.

FIG. 1 shows an environment within which a method and a system for service load distribution in a data network can be implemented, according to an example embodiment.

FIG. 2 is a block diagram showing various modules of a system for service load distribution in a data network.

FIG. 3 is a scheme for service load distribution of a data network.

FIG. 4 is a scheme showing graceful scaling in a data network.

FIG. 5 is a scheme showing graceful scaling in a data network.

FIG. 6 is a scheme showing graceful scaling out a data network.

FIG. 7 is a process flow diagram showing a method for graceful scaling in a data network.

FIG. 8 is a process flow diagram showing a method for graceful scaling out a data network.

FIG. 9 shows a diagrammatic representation of a computing device for a machine in the example electronic form of a computer system, within which a set of instructions for causing the machine to perform any one or more of the methodologies discussed herein, can be executed.

DETAILED DESCRIPTION

The following detailed description includes references to the accompanying drawings, which form a part of the detailed description. The drawings show illustrations in accordance with example embodiments. These example embodiments, which are also referred to herein as “examples,” are described in enough detail to enable those skilled in the art to practice the present subject matter. The embodiments can be combined, other embodiments can be utilized, or structural, logical, and electrical changes can be made without departing from the scope of what is claimed. The following detailed description is therefore not to be taken in a limiting sense, and the scope is defined by the appended claims and their equivalents. In this document, the terms “a” and “an” are used, as is common in patent documents, to include one or more than one. In this document, the term “or” is used to refer to a nonexclusive “or,” such that “A or B” includes “A but not B,” “B but not A,” and “A and B,” unless otherwise indicated.

The present disclosure relates to scaling in a data network, such as an SDN. The techniques of the embodiments disclosed herein may be implemented using a variety of technologies. For example, the methods described herein are implemented in software executing on a computer system or in hardware utilizing either a combination of microprocessors or other specially designed application-specific integrated circuits (ASICs), programmable logic devices, or various combinations thereof. In particular, the methods described herein may be implemented by a series of computer-executable instructions residing on a storage medium such as a disk drive or computer-readable medium. It should be noted that methods disclosed herein can be implemented by a computer (e.g., a desktop computer, a tablet computer, a laptop computer, and a server), game console, handheld gaming device, cellular phone, smart phone, smart television system, and so forth.

According to an example embodiment, the method for graceful scaling a data network comprises receiving an indication of removing a first node from a plurality of nodes of the data network, generating a service policy, sending the service policy to each of the plurality of nodes of the data network, wherein the service policy reassigns service requests associated with the first node to a second node from the plurality of nodes of the data network. In some embodiments, the method for graceful scaling a data network comprises receiving an indication of a presence of a further node in the data network, generating a further service policy, and sending the further service policy to each of the plurality of nodes of the data network and to the further node, wherein the further service policy reassigns one or more of the service requests associated with any of the plurality of nodes of the data network to the further node.

Referring now to the drawings, FIG. 1 illustrates an environment 100 within which a method and a system for load distribution in an SDN can be implemented. The environment 100 includes a network 110, a client 120, a system 200 for load distribution, and servers 140. The client 120 includes a user or a host associated with the network 110.

The network 110 includes the Internet or any other network capable of communicating data between devices. Suitable networks include an interface with any one or more of, for instance, a local intranet, a PAN (Personal Area Network), a LAN (Local Area Network), a WAN (Wide Area Network), a MAN (Metropolitan Area Network), a virtual private network (VPN), a storage area network (SAN), a frame relay connection, an Advanced Intelligent Network (AIN) connection, a synchronous optical network (SONET) connection, a digital T1, T3, E1 or E3 line, Digital Data Service (DDS) connection, DSL (Digital Subscriber Line) connection, an Ethernet connection, an ISDN (Integrated Services Digital Network) line, a dial-up port such as a V.90, V.34 or V.34 bis analog modem connection, a cable modem, an ATM (Asynchronous Transfer Mode) connection, or an FDDI (Fiber Distributed Data Interface) or CDDI (Copper Distributed Data Interface) connection. Furthermore, communications may also include links to any of a variety of wireless networks, including WAP (Wireless Application Protocol), GPRS (General Packet Radio Service), GSM (Global System for Mobile Communication), CDMA (Code Division Multiple Access) or TDMA (Time Division Multiple Access), cellular phone networks, GPS (Global Positioning System), CDPD (cellular digital packet data), RIM (Research in Motion, Limited) duplex paging network, Bluetooth radio, or an IEEE 802.11-based radio frequency network. The network 110 can further include or interface with any one or more of an RS-232 serial connection, an IEEE-1394 (Firewire) connection, a Fiber Channel connection, an IrDA (infrared) port, a SCSI (Small Computer Systems Interface) connection, a USB (Universal Serial Bus) connection or other wired or wireless, digital or analog interface or connection, mesh or Digi® networking. The network 110 may include a network of data processing nodes that are interconnected for the purpose of data communication. The network 110 includes an SDN. The SDN includes one or more of the above network types. Generally, the network 110 includes a number of similar or dissimilar devices connected together by a transport medium enabling communication between the devices by using a predefined protocol. Those skilled in the art will recognize that the present disclosure can be practiced within a variety of network configuration environments and on a variety of computing devices.

As shown in FIG. 1, the client 120 sends service requests 150 to backend servers 140. The service requests 150 include an HTTP request, a video streaming request, a file download request, a transaction request, a conference request, and so forth. The servers 140 include a web server, a wireless application server, an interactive television server, and so forth. The system 200 for load distribution balance flow of the service requests 150 among traffic forwarding devices of the network 110. The system 200 for load distribution may analyze the flow of the service requests 150 and determine which and how many traffic forwarding devices of the network 110 are needed to deliver the service requests 150 to the servers 140.

FIG. 2 shows a block diagram illustrating various modules of a system 200 for service load distribution in an SDN, according to an example embodiment. The system 200 includes a cluster of devices eligible as a cluster master. The system 200 includes a cluster master 205 elected from these devices. The cluster master 205 is configured to keep track of the SDN and retrieve network data associated with the SDN. In an example embodiment, the network data include one or more of the following: a number of total connections, processing unit utilization, a memory status, a network connectivity, backend server capacity, and so forth. Furthermore, the cluster master 205 is configured to keep track of the service nodes and retrieve service node data associated with one or more service nodes. The service node data include one or more of the following: health, dynamic state, responsiveness of the one or more service nodes, and so forth. In other words, the cluster master 205 keeps track of the health of the network and each service node associated with the system 200. The cluster master 205 may analyze the retrieved network data and service node data. Based on the analysis, the cluster master 205 generates a service policy. The service policy includes a service address, a service node address, a service node load policy, a traffic distribution policy (also referred to as a traffic mapping), and so forth. The cluster master 205 may provide the generated service policy to the devices associated with the data network, such as service nodes and traffic classification engines.

In an example embodiment, the cluster master 205 is further configured to develop, based on the analysis, a further service policy. The further policy is associated with scaling out; scaling down; remedying; removing devices, such as service nodes, traffic classification engines, backend servers and so forth; and introducing new service nodes, traffic classification engines, backend servers, and so forth.

In an example embodiment, the cluster master 205 is further configured to facilitate an application programmable interface (not shown) for a network administrator to enable the network administrator to develop, based on the analysis, a further service policy using the retrieved network data and service node data and analytics. This approach allows application developers to write directly to the network without having to manage or understand all the underlying complexities and subsystems that compose the network.

In a further example embodiment, the cluster master 205 may include a backup unit (not shown) configured to replace the cluster master in case of a failure of the cluster master 205.

The system 200 may comprise a traffic classification engine 210. The traffic classification engine 210 may be implemented as one or more software modules, hardware modules, or a combination of hardware and software. The traffic classification engine 210 may include an engine configured to monitor data flows and classify the data flows based on one or more attributes associated with the data flows (e.g., uniform resource locators (URLs), IP addresses, port numbers, and so forth). Each resulting data flow class can be specifically designed to implement a certain service for a client. In an example embodiment, the cluster master 205 may send a service policy to the traffic classification engine 210. The traffic classification engine 210 may be configured to receive the service policy from the cluster master 205. Furthermore, the traffic classification engine 210 may be configured to receive one or more incoming service requests 215 (e.g. incoming data traffic from routers or switches (not shown)). Typically, the data traffic may be distributed from the routers or switches to each of the traffic classification engines 210 evenly. In an example embodiment, a router performs a simple equal-cost multi-path (ECMP) routing to distribute the traffic equally to all the traffic classification engines 210. The traffic classification engines 210 distribute the one or more service requests among one or more service nodes 220 according to the service policy. The traffic is distributed to the one or more service nodes 220 in an asymmetric fashion. The traffic to the service nodes 220 may be direct or through a tunnel (IP-in-IP or other overlay techniques). The traffic classification engine 210 may be stateless or stateful, may act on a per packet basis, and may direct each packet of the traffic to the corresponding service node 220. When there is a change in the state of the service nodes, the cluster master 205 sends a new service policy, such as a new traffic map, to the traffic classification engine 210.

The system 200 may comprise the one or more service nodes 220. The one or more service nodes 220 may include a virtual machine or a physical device that may serve a corresponding virtual service to which the traffic is directed. The cluster master 205 sends the service policy to the service nodes 220. The service nodes 220 may be configured to receive the service policy from the cluster master 205. Furthermore, the service nodes 220 receive, based on the service policy, the one or more service requests 215 from the traffic classification engine 210. The one or more service nodes 220 may process the received one or more service requests 215 according to the service policy. The processing of the one or more service requests 215 may include forwarding the one or more service requests 215 to one or more backend destination servers (not shown). Each service node 220 may serve one or more virtual services. The service nodes 220 may be configured to send the service node data to the cluster master 205.

According to further example embodiments, an existing service node may redirect packets for existing flows to another service node if it is the new owner of the flow based on the redistribution of flows to the service nodes. In addition, a service node taking over the flow may redirect packets to the service node that was the old owner for the flows under consideration, for cases where the flow state needs to be pinned down to the old owner to maintain continuity of service.

Furthermore, in an example embodiment, the cluster master 205 may perform a periodic health check on the service nodes 220 and update the service nodes 220 with a service policy, such as a traffic map. When there is a change in the traffic assignment and a packet of the data traffic in a flow reaches a service node 220, the service node 220 may redirect the packet to another service node. Redirection may be direct or through a tunnel (e.g., IP-in-IP or other overlay techniques).

It should be noted that if each of the devices of the cluster in the network performs the backend server health check, it may lead to a large number of health check packets sent to an individual device. In view of this, a few devices of the cluster may perform the backend server health check, and the result may be shared among the rest of the devices in the cluster. The health check may include a service check and a connectivity check. The service check may include determining whether the application or the backend server is still available. As already mentioned above, not every device in the cluster needs to perform this check. The check can be performed by a few devices, and the result propagated to the rest of the devices in the cluster. A connectivity check includes determining whether the service node can reach the backend server. The path to the backend server may be specific to the service node, so this may not be distributed across service nodes, and each device in the cluster may perform its own check.

In an example embodiment, the system 200 comprises an orchestrator 225. The orchestrator 225 may be configured to bring up and bring down the service nodes 220, the traffic classification engines 210, and backend servers. The orchestrator 225 may detect a presence of the one or more service nodes 220 and transmit data associated with the presence of the one or more service nodes 220 to the cluster master 205. Furthermore, the orchestrator 225 may inform the cluster master 205 of bringing up or bringing down the service nodes 220. The orchestrator 225 may communicate with the cluster master 205 and the service nodes 220 using one or more Application Programming Interfaces (APIs).

In an example embodiment, a centralized or distributed network database may be used and shared among all devices in the cluster of the system 200, such as the cluster master, the traffic classification engine, and other service nodes. Each device may connect to the network database and update tables according to its role. Relevant database records may be replicated to the devices that are part of the cluster. The distributed network database may be used to store configurations and states of the devices (e.g., to store data associated with the cluster master, the traffic classification engine, the one or more service nodes, and backend servers). The data stored in the distributed network database may include the network data and the service node data. The distributed network database may include tables with information concerning service types, availability of resources, traffic classification, network maps, and so forth. The cluster master 205 may be responsible for maintaining the distributed network database and replicating it to devices. The network database may be replicated to the traffic classification engines 210 and the service nodes 220. In an example embodiment, the network database may internally replicate data across the participant nodes.

In the embodiments described above, the system 200 comprises a dedicated cluster master 205, dedicated traffic classification engines 210, and dedicated service nodes 220. In other words, specific devices can be responsible for acting as the cluster master, the traffic classification engine, and the service node. In further example embodiments, the system 200 includes no dedicated devices acting as a cluster master. In this case, the cluster master functionality is provided by either the traffic classification engines or by the service nodes. Thus, one of the traffic classification engines or one of the service nodes is operable to act as the cluster master. In case the traffic classification engine or service node acting as the cluster master fails, another traffic classification engine or service node may be elected as the cluster master. The traffic classification engines and the service nodes not elected as the cluster master are configured as backup cluster masters and synchronized with the current cluster master. In an example embodiment, the cluster master consists of multiple active devices that can act as a single master by sharing duties among the devices.

In further example embodiments, the system 200 comprises a dedicated cluster master with no dedicated devices acting as traffic classification engines. In this case, the traffic classification may be performed by one of upstream routers or switches. Also, the service nodes may distribute the traffic among themselves. In an example embodiment, the cluster master and the service nodes are configured to act as a traffic classification engine.

In further example embodiments, the system 200 includes no devices acting as cluster masters and traffic classification engines. In this case, one of the service nodes is configured to also act as the cluster master. Upstream routers or switches can do the traffic classification. The cluster master programs the upstream routers with the traffic mapping. Additionally, the service nodes distribute the traffic among themselves.

It should be noted that bringing up new service nodes when the load increases and bringing down the service nodes when the load becomes normal can be performed gracefully, without affecting existing data traffic and connections. When the service node comes up, the distribution of traffic changes from distribution to n service nodes to distribution to (n+1) service nodes.

When a service node is about to be brought down, the traffic coming to this service node is redirected to other service nodes. For this purpose, a redirection policy associated with the service node about to be brought down may be created by the cluster muster and sent to the traffic distribution engine and/or the service nodes. Upon receiving the redirection policy, the traffic distribution engine directs the traffic to another service node.

In an example embodiment, the system 200 comprises, for example, a plurality of traffic distribution engines, each of which serves traffic to multiple services. Each of the traffic distribution engines may communicate with a different set of service nodes. In case one of the traffic distribution engines fails, another traffic distribution engine is configured to substitute the failed traffic distribution engine and to distribute the traffic of the failed traffic distribution engine to the corresponding service nodes. Therefore, each of the traffic distribution engines comprises addresses of all service nodes and not only addresses associated with the service nodes currently in communication with the traffic distribution engine.

FIG. 3 shows a block diagram 300 for load distribution of an SDN, according to an example embodiment. As shown, diagram 300 includes clients 120 (e.g., a computer connected to a network 110). The network 110 may include the SDN. The clients 120 may send one or more service requests for services provided by one or more servers of a virtual machine/server pool 305. These servers include web servers, wireless application servers, interactive television servers, and so forth. These service requests can be load balanced by a system for load distribution described above. In other words, the service requests of the clients 120 may be intelligently distributed among virtual machine/server pool 305 of the SDN.

The system for load distribution includes a service control plane 310. The service control plane 310 includes one or more data network applets 315 (for example, a real time data network applet). The data network applets 315 check the health and other data associated with the SDN 110 and the virtual machines 305. For example, the data network applets 315 may determine responsiveness of the virtual machine/server pool 305. Furthermore, the data network applets 315 monitor the total connections, central processing unit utilization, memory, network connectivity on the virtual machine/server pool 305, and so forth. Therefore, the data network applets 315 may retrieve fine-grained, comprehensive information concerning the SDN and virtual machine service infrastructure.

The retrieved health data may be transmitted to a service policy engine 320. In example embodiments, a cluster master 205 described above may act as the service policy engine 320. The service policy engine 320 may analyze the health data and, upon the analysis, generate a set of service policies 330 to scale up/down the services, to secure services, to introduce new services, to remove services, to remedy or repair failed devices, and so forth. The system for load distribution may further comprise an orchestrator (not shown) configured to bring up more virtual machines on demand. Therefore, in order to deliver a smooth client experience, the service requests may be load balanced across the virtual machine/server pool 305.

Furthermore, service policies 330 may be provided to an SDN controller 335. The SDN controller 335, in turn, may steer service requests, i.e., data traffic, across the network devices in the SDN. Effectively, these policies may control load balancing, availability, and functionality of the SDN network to scale up or scale down services.

Generally speaking, by unlocking the data associated with the network, service nodes, and the server/virtual machines from inside the network, transforming the data into relevant information and the service policies 330, and then presenting the service policies 330 to the SDN controller 335 for configuring the SDN 110, the described infrastructure may enable feedback loops between underlying infrastructure and applications to improve network optimization and application responsiveness.

The service control plane 310 working in conjunction with the controller 335 and the service policy engine 320 may create a number of deployment possibilities, which may offer an array of basic and advanced load distribution features. In particular, to provide a simple load balancing functionality, the SDN controller 335 and the service control plane 310 may provide some load balancing of their own by leveraging the capabilities of the SDN 110 or, alternatively, work in conjunction with an ADC 340, also referred to as a service data plane included in the SDN 110 to optionally provide advanced additional functionality.

In an example embodiment, the service control plane 310 may be standalone, i.e., without an ADC 340, virtual machines 305 or, when scaled up, may be programmed with a virtual Internet Protocol (VIP) address on a loopback interface of the virtual machines 305. Thus, for data traffic in need of simple service fulfillment, the service control plane 310 may establish simple policies for distributing service requests and instruct the SDN controller 335 to program network devices to distribute the service requests directly to different virtual machines/physical servers 305. This step may be performed over a physical or logical network.

In an example embodiment, while the service control plane 310 may work in cooperation with an ADC 340, the service control plane 310 may manage a set of service policy mapping service requests to one or more ADC devices for more sophisticated ADC functionality typically offered by a purpose built ADC device. The service control plane 310 may instruct the SDN controller 335 to program network devices such that the service requests, i.e., the traffic, may reach one or more ADCs 340. The ADC 340 then may relay the service request to a backend server over a physical or logical network.

In the described embodiment, several traffic flow scenarios may exist. In an example embodiment, only forward traffic may go through the ADC 340. If a simple functionality of the ADC 340 (e.g., rate limiting, bandwidth limiting, scripting policies) is needed, the forward traffic may traverse the ADC 340. The loopback interface on the servers may be programmed with the VIP address. Response traffic from the virtual machines 305 may bypass the ADC 340.

In a further example embodiment, forward and reverse traffic may traverse the ADC 340. If it is needed that the ADC 340 provide a more advanced functionality (e.g., transmission control protocol (TCP) flow optimization, secure sockets layer (SSL) decryption, compression, caching and so forth), the service control plane 310 may need to ensure that both the forward and reverse traffic traverses through the ADC 340 by appropriately programming the SDN 110.

FIG. 4 is a scheme 400 showing graceful scaling in a data network, according to an example embodiment. In example scheme 400, two service nodes 420 (B1) and 425 (B2) receive data traffic from node 410 (D1) belonging to traffic classification engine 210 (shown in FIG. 2). The service node 425 is being removed from the data network. The node 410 sends the data traffic according to a service policy described in traffic map 440 generated by a cluster master 205 (shown in FIG. 2).

In the example traffic map 440, the column A contains “active” nodes and the column S contains the “stand by” nodes for service requests belonging to class C1 and C2. According to the traffic map 440, the all service requests belonging to class C1 are sent first to the node B1 and, if node B1 cannot handle the received service requests, to the node B2. Correspondingly, all service requests classified to class C2 are sent first to the node B2, and, if node B2 cannot handle the received packet, to the node B1.

In order to remove the service node B2, the cluster master 205 first generates new traffic map 450, wherein the node B2 is removed from the “stand by” column. Another service node from the data network, for example Bk, may be assigned as the “stand by” node for the packets belonging to class C1.

Further, the cluster master generates another traffic map 460, wherein the node B2 is replaced as “active” node by the “stand by” node for the corresponding class of the service requests. In example scheme 400, the B2 node is replaced by the B1 node. The cluster master sends the copy of the new traffic map 460 to the nodes of the traffic classification engine and service nodes of the data network. The B2 node can be safely removed or shutdown from the data network.

It should be noted that example scheme 400 for graceful scaling in a data network is suitable for removing nodes involved in connections with light traffic flow.

FIG. 5 is a scheme 500 showing graceful scaling in a data network, according to another example embodiment. The example scheme 500 is suitable for removing nodes involved in heavy duty traffic connections (for example, SSL (Secure Sockets Layer) type connections.)

In example scheme 500, two service nodes 420 (B1) and 425 (B2) receive data traffic from node 410 (D1) belonging to traffic classification engine 210 (shown in FIG. 2). The node 410 sends the data traffic according to a service policy described in traffic map 440 generated by a cluster master 205 (shown in FIG. 2). The service node 425 is needed to be removed from the data network.

In order to remove the B2 node, the cluster master generates and sends to the traffic classification engine and service nodes new traffic map 460 as described in FIG. 4. Before generating and sending the new traffic map 460, an indication of a presence of and old connections associated with the B2 node is received, and a redirection node 520 (R1) is created. The cluster master 205 sends a redirection service policy to the nodes B1 and R1. According to the redirection service policy, all service requests belonging to the old connections should be sent from the node B1 through the redirection node R1 to the B2 node. All other service requests flow through the B1 node according to the new traffic map 460. After the cluster map receives an indication from the service node B2 that all old connections are finished, the redirection node R1 may be removed from data network and the redirection service policy can be cleared from the B1 node. The B2 service node can be safely removed from the data network.

FIG. 6 is a scheme 600 showing graceful scaling out a data network, according to an example embodiment. In example scheme 600 the service node 420 (B1) receives data traffic from node 410 (D1) belonging to traffic classification engine 210 (shown in FIG. 2). The node 410 sends the data traffic according to a service policy described in traffic map 640 generated by a cluster master 205 (shown in FIG. 2). The service node 425 (B2) is added to the data network.

In the example traffic map 640, the column A contains “active” nodes and the column S contains the “stand by” nodes for service requests belonging to class C1 and C2. According to the traffic map 640, all service requests belonging to classes C1 and C2 are send first to the node B1 and, if node B1 cannot handle the received service requests, then the packets of class C1 are sent to the node Bk and the packets of class C2 are sent to node Bn.

After service node B2 becomes active in the data network, the cluster master may generate new traffic map 660. According to new traffic map 660, all service requests classified to class C2 should be sent to the node B2. The cluster master sends the new traffic map to nodes of traffic classification engine and all service nodes in the data network.

If the service node B1 was handling a heavy traffic connection before adding new node B2, such as a SSL connection, then before the cluster master generates and sends the new traffic map 660, a redirection node 620 (R2) can be created. The cluster master sends a redirection service policy to the node B2 and R2. According to the redirection service policy, all service requests belonging to old connections that flowed through the B1 node before generating the new traffic map 460 should be sent from the B2 node through the redirection node R2 to the B1 node. All other service requests flow through the B2 node according to the service policy described in the new traffic map 460. After the cluster map receives an indication from the service node B1 that all old connections are finished, the redirection node R2 may be removed from data network and the redirection service policy can be cleared from the B2 node.

FIG. 7 is a process flow diagram showing a method 700 for graceful scaling in a data network, according to an example embodiment. The method 700 can be performed by processing logic that may comprise hardware (e.g., decision making logic, dedicated logic, programmable logic, and microcode), software (such as software running on a general-purpose computer system or a dedicated machine), or a combination of both. In particular, the method 700 can be implemented in example SDN shown in FIG. 2.

The method 700 may commence in step 702 with receiving an indication of removing a node from a plurality of nodes in a data network. In step 704, the method 700 proceeds further with generation of a service policy, and the service policy reassigns service requests associated with the first node to a second node from the plurality of nodes in the data network.

The method 700 may proceed further with following optional steps 706-712. In optional step 706, the method 700 continues with receiving an indication of a presence of old connections associated with the first node. In step 708, a redirection node is created in data network. In step 710, a redirection policy is generated. The redirection policy contains an indication that service requests belonging to old connections should be sent from the second node via the redirection node to the first node. In step 712, the redirection policy is sent to the second node and the redirection node.

In step 714, the service policy generated in step 704 is sent to each node in the plurality of nodes in the data network. The method 700 can further proceed with following optional steps 716-720. In optional 716, the method 700 proceeds with waiting for an indication that all old connections in the first node are finished. In step 718, the redirection policy is removed from the second node. In step 720, the redirection node is removed from the data network.

FIG. 8 is a process flow diagram showing a method 800 for graceful scaling out a data network, according to an example embodiment. The method 800 can be performed by processing logic that may comprise hardware (e.g., decision making logic, dedicated logic, programmable logic, and microcode), software (such as software running on a general-purpose computer system or a dedicated machine), or a combination of both. In particular, the method 800 can be implemented in example SDN shown in FIG. 2.

The method 800 may commence in step 802 with receiving an indication of a presence of a further node in a plurality of nodes of a data network. In step 804, a further node policy is generated. The further node policy reassigns to the further node one or more service requests associated earlier with any of plurality of nodes of the data network.

The method 800 may further proceed with optional steps 806-812. In optional step 806, the method 800 may proceed with receiving an indication of old connections associated with an old node of the plurality of the nodes of the data network. In step 808, a redirection node can be created in data network. In step 810, a redirection policy is generated. The redirection policy indicates that service requests associated with old connections should be sent to the further node via the redirection node to the old node. In step 812, the redirection policy is sent to the redirection node and to the further node.

In step 814, the further node policy is sent to all nodes in the plurality of nodes of the data network and to the further node.

The method 800 may further proceed with optional steps 816-820. In step 816, the method 800 continues with waiting for an indication that all old connections are finished. In step 818, the redirection policy is removed from the further node. In step 820, the redirection node is removed from the data network.

FIG. 9 shows a diagrammatic representation of a machine in the example electronic form of a computer system 900, within which a set of instructions for causing the machine to perform any one or more of the methodologies discussed herein may be executed. In various example embodiments, the machine operates as a standalone device or may be connected (e.g., networked) to other machines. In a networked deployment, the machine may operate in the capacity of a server or a client machine in a server-client network environment, or as a peer machine in a peer-to-peer (or distributed) network environment. The machine may be a personal computer (PC), a tablet PC, a set-top box (STB), a cellular telephone, a portable music player (e.g., a portable hard drive audio device such as an Moving Picture Experts Group Audio Layer 3 (MP3) player), a web appliance, a network router, switch or bridge, or any machine capable of executing a set of instructions (sequential or otherwise) that specify actions to be taken by that machine. Further, while only a single machine is illustrated, the term “machine” shall also be taken to include any collection of machines that individually or jointly execute a set (or multiple sets) of instructions to perform any one or more of the methodologies discussed herein.

The example computer system 900 includes a processor or multiple processors 902 (e.g., a central processing unit (CPU), a graphics processing unit (GPU), or both), a main memory 904, and a static memory 906, which communicate with each other via a bus 908. The computer system 900 may further include a video display unit 910 (e.g., a liquid crystal display (LCD) or a cathode ray tube (CRT)). The computer system 900 may also include an alphanumeric input device 912 (e.g., a keyboard), a cursor control device 914 (e.g., a mouse), a disk drive unit 916, a signal generation device 918 (e.g., a speaker), and a network interface device 920.

The disk drive unit 916 includes a non-transitory computer-readable medium 922, on which is stored one or more sets of instructions and data structures (e.g., instructions 924) embodying or utilized by any one or more of the methodologies or functions described herein. The instructions 924 may also reside, completely or at least partially, within the main memory 904 and/or within the processors 902 during execution thereof by the computer system 900. The main memory 904 and the processors 902 may also constitute machine-readable media.

The instructions 924 may further be transmitted or received over a network 926 via the network interface device 920 utilizing any one of a number of well-known transfer protocols (e.g., HTTP).

While the computer-readable medium 922 is shown in an example embodiment to be a single medium, the term “computer-readable medium” should be taken to include a single medium or multiple media (e.g., a centralized or distributed database and/or associated caches and servers) that store the one or more sets of instructions. The term “computer-readable medium” shall also be taken to include any medium that is capable of storing, encoding, or carrying a set of instructions for execution by the machine and that causes the machine to perform any one or more of the methodologies of the present application, or that is capable of storing, encoding, or carrying data structures utilized by or associated with such a set of instructions. The term “computer-readable medium” shall accordingly be taken to include, but not be limited to, solid-state memories, optical and magnetic media, and carrier wave signals. Such media may also include, without limitation, hard disks, floppy disks, flash memory cards, digital video disks (DVDs), random access memory (RAM), read only memory (ROM), and the like.

The example embodiments described herein can be implemented in an operating environment comprising computer-executable instructions (e.g., software) installed on a computer, in hardware, or in a combination of software and hardware. The computer-executable instructions can be written in a computer programming language or can be embodied in firmware logic. If written in a programming language conforming to a recognized standard, such instructions can be executed on a variety of hardware platforms and for interfaces to a variety of operating systems. Although not limited thereto, computer software programs for implementing the present method can be written in any number of suitable programming languages such as, for example, Hypertext Markup Language (HTML), Dynamic HTML, Extensible Markup Language (XML), Extensible Stylesheet Language (XSL), Document Style Semantics and Specification Language (DSSSL), Cascading Style Sheets (CSS), Synchronized Multimedia Integration Language (SMIL), Wireless Markup Language (WML), Java™, Jini™, C, C++, Perl, UNIX Shell, Visual Basic or Visual Basic Script, Virtual Reality Markup Language (VRML), ColdFusion™ or other compilers, assemblers, interpreters or other computer languages or platforms.

Thus, methods and systems for graceful scaling nodes in an SDN are disclosed. Although embodiments have been described with reference to specific example embodiments, it will be evident that various modifications and changes can be made to these example embodiments without departing from the broader spirit and scope of the present application. Accordingly, the specification and drawings are to be regarded in an illustrative rather than a restrictive sense.

Claims

1. A computer-implemented method for graceful scaling of a data network, the method comprising: receiving, by a cluster master, a first indication of a removal of a first node from a plurality of nodes of the data network;receiving, by the cluster master, a second indication of a presence of at least one old connection associated with the first node, the at least one old connection including at least one unfinished connection currently served by the first node;in response to receiving the first indication and the second indication, generating, by the cluster master, a service policy, the service policy being operable to reassign service requests associated with the first node to a second node selected from the plurality of nodes of the data network, the service policy indicating that one or more of the service requests associated with the at least one old connection are to be redirected by the second node to the first node, the service requests including requests sent by one or more clients for services provided by one or more of the plurality of nodes;sending, by the cluster master, the service policy to one or more of the plurality of nodes of the data network;receiving, by the cluster master, from the first node, an indication of a completion of the at least one old connection associated with the first node; andremoving the first node from the data network upon the completion of the at least one old connection associated with the first node.

2. The method of claim 1, wherein the plurality of nodes of the data network include a virtual machine and a physical device.

3. The method of claim 1, wherein the plurality of nodes include a plurality of cluster devices, the cluster master, a traffic classification engine, service nodes, and an orchestrator.

4. The method of claim 3, wherein the traffic classification engine is configured to distribute the service requests processed by the data network to the service nodes in accordance with the service policy.

5. The method of claim 1, wherein the first indication is generated based on one or more of the following: first node health, a first node service policy, analysis of a first node data, and absence of the first node data.

6. The method of claim 5, wherein the first node data includes a number of total connections, destination server address, destination server capacity, and destination server network connectivity.

7. The method of claim 1, wherein the service policy includes a traffic map table, wherein classes of the service requests are assigned to nodes in the plurality of nodes of the data network.

8. The method of claim 7, wherein the classes of the services requests are obtained by classification of the service request based on one of the following: a source address of the service request and a destination address of the service request.

9. The method of claim 1, further comprising, before sending the service policy to each of the plurality of nodes of the data network: creating, by the cluster master, a redirection node in the data network;generating, by the cluster master, a redirection policy, the redirection policy indicating that service requests associated with the old connections are to be sent from the second node to the first node via the redirection node; andsending, by the cluster master, the redirection policy to the redirection node and the second node; andafter sending the service policy to each of the plurality of nodes of the data network:waiting for an indication of finishing the old connections and upon receiving the indication of finishing the old connections: removing, by the cluster master, the redirection policy from the second node; andremoving, by the cluster master, the redirection node from the data network.

10. The method of claim 1, further comprising: receiving, by the cluster master, an indication of presence of a further node in the data network;generating, by the cluster master, a further node service policy, the further node service policy reassigning one or more of the service requests associated with any of the plurality of nodes of the data network to the further node; andsending, by the cluster master, the further node service policy to each of the plurality of nodes of the data network and to the further node.

11. The method of claim 10, further comprising before sending the further node service policy to each of the plurality of nodes of the data network: receiving, by the cluster master, an indication of presence of old connections, the old connections being associated with an old node from the plurality of nodes of the data network;creating, by the cluster master, a redirection node in the data network;generating, by the cluster master, a redirection policy, the redirection policy indicating that service requests associated with the old connections be sent from the further node via the redirection node to the old node; andsending, by the cluster master, the redirection policy to the redirection node and the further node; andafter sending the further node service policy to each of the plurality of nodes of the data network and to the further node:waiting for an indication of finishing the old connections and upon receiving the indication of finishing the old connections: removing, by the cluster master, the redirection policy from the further node; andremoving, by the cluster master, the redirection node from the data network.

12. A system for graceful scaling a data network, the system comprising: a plurality of nodes; anda cluster master, the cluster master operable to: receive a first indication of removing a first node from the plurality of nodes of the data network;receive a second indication of a presence of at least one old connection associated with the first node, the at least one old connection including at least one unfinished connection currently served by the first node;in response to the first indication and the second indication, generate a service policy, the service policy reassigning any service requests associated with the first node to a second node from the plurality of nodes of the data network, the service policy indicating that one or more of the service requests associated with the at least one old connection are to be redirected by the second node to the first node, the service requests including requests sent by one or more clients for services provided by one or more of the plurality of nodes;send the service policy to each of the plurality of nodes of the data network;receive from the first node, an indication of a completion of the at least one old connection associated with the first node; andremove the first node from the data network upon the completion of the at least one old connection associated with the first node.

13. The system of claim 12, wherein the plurality of nodes of the data network include a virtual machine and a physical device.

14. The system of claim 12, wherein the plurality of nodes include a plurality of cluster devices, a traffic classification engine, service nodes, and an orchestrator.

15. The system of claim 12, wherein the first indication is generated based on one or more of the following: a first node health, a first node service policy, an analysis of the first node data, and an absence of the first node data.

16. The system of claim 15, wherein the first node data includes a number of total connections, destination server address, destination server capacity, and destination server network connectivity.

17. The system of claim 12, wherein the service policy is a traffic map table, wherein classes of the services requests are assigned to nodes in the plurality of nodes of the data network.

18. The system of claim 17, wherein the classes of the service requests are obtained by classification of the service requests based on one of the following: a source address of the service request and a destination address of the service request.

19. The system of claim 12, wherein before sending the service policy to each of the plurality of nodes of the data network, the cluster master is further operable to: receive an indication of presence of old connections, the old connections being associated with the first node;generate an instruction to create a redirection node in the data network;generate a redirection policy, the redirection policy indicating that service requests associated with the old connections be sent from the second node to the first node via the redirection node; andsend the redirection policy to the redirection node and the second node; andafter sending the service policy to each of the plurality of nodes of the data network, the cluster master is further operable to:wait for an indication of finishing the old connections and upon receiving the indication of finishing the old connections: remove the redirection policy from the second node; andgenerate an instruction to remove the redirection node from the data network.

20. The system of claim 12, wherein the cluster master is further operable to: receive an indication of presence of a further node in the data network;generate a further node service policy, the further node service policy reassigning one or more of the service requests associated with any of the plurality of nodes of the data network to the further node; andsend the further node service policy to each of the plurality of nodes of the data network and to the further node.

21. The system of claim 20, wherein before sending the further node service policy to each of the plurality of nodes of the data network, the cluster master is further operable to: receive an indication of presence of old connections, the old connections being associated with an old node from the plurality of nodes of the data network;generate an instruction to create a redirection node in the data network;generate a redirection policy, the redirection policy indicating that service requests associated with the old connections be sent from the further node via the redirection node to the old node; andsend the redirection policy to the redirection node and the further node; andafter sending the further node service policy to each of the plurality of nodes of the data network and to the further node, the cluster master is further operable to:wait for an indication of finishing the old connections and upon receiving the indication of finishing the old connections: remove the redirection policy from the further node; andgenerate an instruction to remove the redirection node from the data network.

22. A non-transitory computer-readable storage medium having embodied thereon a program, the program being executable by a processor to perform a method for graceful scaling of a data network, the method comprising: receiving, by a cluster master, a first indication of removing a first node from a plurality of nodes of the data network;receiving, by the cluster master, a second indication of a presence of old connections, the old connections being associated with the first node, the at least one old connection including at least one unfinished connection currently served by the first node;in response to receiving the first indication and the second indication, generating, by the cluster master, a service policy, the service policy reassigning any service requests associated with the first node to a second node from the plurality of nodes of the data network, the service policy indicating that one or more of the service requests associated with the at least one old connection are to be redirected by the second node to the first node, the service requests including requests sent by one or more clients for services provided by one or more of the plurality of nodes;sending the service policy to each of the plurality of nodes of the data network;receiving, by the cluster master, from the first node, an indication of a completion of the at least one old connection associated with the first node; andremoving the first node from the data network upon the completion of the at least one old connection associated with the first node.

23. The non-transitory computer-readable storage medium of claim 22, wherein the method for graceful scaling of the data network further comprises before sending the service policy to each of the plurality of nodes of the data network: creating, by the cluster master, a redirection node in the data network;generating, by the cluster master, a redirection policy, the redirection policy indicating that service requests associated with the old connections be sent from the second node to the first node via the redirection node; andsending, by the cluster master, the redirection policy to the redirection node and the second node; andafter sending the service policy to each of the plurality of nodes of the data network:waiting for an indication of finishing the old connections and upon receiving the indication of finishing the old connections: removing, by the cluster master, the redirection policy from the second node; andremoving, by the cluster master, the redirection node from the data network.

24. The non-transitory computer-readable storage medium of claim 22, wherein the method for graceful scaling of the data network further comprises: receiving, by the cluster master, an indication of presence of a further node in the data network;generating, by the cluster master, a further node service policy, the further node service policy reassigning one or more of the service requests associated with any of the plurality of nodes of the data network to the further node; andsending, by the cluster master, the further node service policy to each of the plurality of nodes of the data network and to the further node.

25. The non-transitory computer-readable storage medium of claim 24, wherein the method for graceful scaling of the data network further comprises before sending the further node service policy to each of the plurality of nodes of the data network: receiving, by a cluster master, an indication of presence of old connections, the old connections being associated with an old node from the plurality of nodes of the data network;creating, by the cluster master, a redirection node in the data network;generating, by the cluster master, a redirection policy, the redirection policy indicating that service requests associated with the old connections be sent from the further node via the redirection node to the old node; andsending, by the cluster master, the redirection policy to the redirection node and the further node; andafter sending the further node service policy to each of the plurality of nodes of the data network and to the further node:waiting for an indication of finishing the old connections and upon receiving the indication of finishing the old connections: removing, by the cluster master, the redirection policy from the further node; andremoving, by the cluster master, the redirection node from the data network.