Method and apparatus for sve redundancy

Abstract

Systems and methods for providing service virtualization endpoint (SVE) redundancy in a two-node, active-standby form. An active-standby pair of SVEs register with a cloud-centric-network control point (CCN-CP) as a single service node (SN) using a virtual IP address for both a control-plane and a data-plane. At any given time, only the active SVE is a host for the control-plane and the data-plane. When a failover happens, the hosting operation is taken over by the standby SVE, therefore the failover will be transparent to CCN-CP and the SN.

1 . A method for providing redundancy in a service insertion architecture, comprising:
 providing a service classifier (SCL) that performs traffic classification and service header insertion, the service header including service ordering information related to one or more service nodes that apply service-specific policies to packets received at the SCL; providing a first services virtualization endpoint (SVE) and a second services virtualization endpoint (SVE), the first SVE and the second SVE each sharing a same virtual IP address in a control path, the first SVE and the second SVE each connected to the one or more service nodes in a data path;   duplicating service chaining information in the first SVE and the second SVE,   redirecting the packets received at the SCL to the virtual IP address;   forwarding the packets, via the virtual IP address, to one of the first SVE and the second SVE in the data path and then to the one or more service nodes in accordance with the service header, and   returning the packets to the SCL.   
     
     
         2 . A method according to  claim 1 , further comprising accessing the service header with one of the first and second SVEs to utilize service ordering information within the service header and related to one or more service nodes that apply service-specific policies to packets received at the SCL. 
     
     
         3 . The method of  claim 2 , further comprising:
 providing a cloud centric network control point (CCN-CP) that maintains the mapping, which is an ordered list of the service nodes and a path connecting the service nodes, and configuring the SCL with a load balancer to manage the redirecting of the packets to the virtual IP address.   
     
     
         4 . The method of  claim 3 , further comprising:
 storing the service chaining information in a ternary content addressable memory (TCAM); and   replicating, by the CCN-CP, the TCAM entries between the first SVE and the second SVE.   
     
     
         5 . The method of  claim 1 , wherein the service nodes communicate to the virtual IP address. 
     
     
         6 . The method of  claim 5 , further comprising directing the packets to the first SVE. 
     
     
         7 . The method of  claim 1 , further comprising:
 storing the service chaining information in a ternary content addressable memory (TCAM); and   replicating the TCAM entries between the first SVE and the second SVE, wherein the first SVE and the second SVE communicate directly with each other.   
     
     
         8 . The method of  claim 1 , further comprising:
 detecting a failure of the first SVE;   redirecting packets received at the SCL to the second SVE at the virtual IP address; and   directing the packets in accordance with the mapping for processing.   
     
     
         9 . The method of  claim 1 , further comprising storing the service chaining information in a ternary content addressable memory (TCAM); and
 replicating predetermined ones of the TCAM entries between the first SVE and the second SVE.   
     
     
         10 . The method of  claim 9 , wherein a RBH (result based hash) is used for load-balancing between the first and second SVEs. 
     
     
         11 . The method of  claim 1 , further comprising:
 storing the service chaining information in a ternary content addressable memory (TCAM); and   replicating the TCAM entries between the first SVE and the second SVE.   
     
     
         12 . The method of  claim 1 , further comprising:
 storing the service chaining information in a ternary content addressable memory (TCAM); and   configuring the first SVE with more TCAM entries than the second SVE.   
     
     
         13 . A service insertion architecture, comprising:
 a cloud-centric-network (CCN) control point that maintains an ordered list of service nodes and a path connecting each element in the order;   a service classifier (SCL) that performs traffic classification and service header insertion, the service header including service ordering information related to one or more service nodes that apply service-specific policies to packets received at the SCL;   a first services virtualization endpoint (SVE) and a second services virtualization endpoint (SVE) the first SVE and the second SVE each sharing a same virtual IP address in a control path, the first SVE and the second SVE each in data communication with the service nodes in a data path, the first SVE and the second SVE each comprising Ternary Content Addressable Memory (TCAM), wherein TCAM entries comprise service chaining information of the first and second SVEs,   wherein said TCAM entries are distributed between the first SVE and the second SVE to effect load balancing of packets between the SVEs.   
     
     
         14 . The service insertion architecture of  claim 13 , wherein one SVE registers with the CCN at the virtual IP address, and wherein a packet enters the service insertion architecture at the SCL and is directed to the service nodes via the one SVE at the virtual IP address, and
 wherein one of the first SVE and the second SVE forwards the packet in the data path to the service nodes in accordance with a mapping comprising an ordered list of the service nodes and a path connecting the service nodes.   
     
     
         15 . The service insertion architecture of  claim 13 , wherein if one of the SVE fails, an other SVE services the packet. 
     
     
         16 . The service insertion architecture of  claim 15 , wherein one of the first and second SVEs is designated a master and another of the SVEs is designated a slave. 
     
     
         17 . The service insertion architecture of  claim 16 , wherein only one of the first and second SVEs is active at any one time in the data plane. 
     
     
         18 . The service insertion architecture of  claim 15 , wherein at least one of the first and second SVEs is implemented in software. 
     
     
         19 . The service insertion architecture of  claim 15 , wherein at least one of the first and second SVEs is implemented in an ASIC. 
     
     
         20 . A service insertion architecture, comprising:
 a cloud centric network control point (CCN-CP) that maintains mapping in both a network node classification context and a service path context;   a service classifier (SCL) that performs traffic classification and service header insertion, the service header including service ordering information related to one or more service nodes that apply service-specific policies to packets received at the SCL;   a first services virtualization endpoint (SVE) and a second services virtualization endpoint (SVE), the first SVE and the second SVEs each sharing a same virtual IP address in a control path, such that the CCN•CP communicates with the virtual IP address to direct packets to one of the first SVE and the second SVE, wherein said first and second SVEs are connected to service nodes in a data path, and wherein only one of the first SVE and the second SVE is active at any one time;   a virtual IP switch interface to the service insertion architecture connecting the service nodes to the virtual IP address connecting the packets to an active SVE.