Method and apparatus for distributing load among a plurality of service nodes

• US 9,825,810 B2
• Filed: 12/12/2014
• Issued: 11/21/2017
• Est. Priority Date: 09/30/2014
• Status: Active Grant

• Current Assignee: Nicira Incorporated (Broadcom, Inc.)
• Original Assignee: Nicira Incorporated (Broadcom, Inc.)
• Inventors: Jain, Jayant, Sengupta, Anirban, Parthasarathy, Mohan
• Primary Examiner(s): Nguyen, Steven H
• Assistant Examiner(s): Lalchinthang, Vanneilian
• Application Number:
  US14/569,405
• Publication Number:
  US 20160094455A1
• Time in Patent Office: 1,075 Days
• Field of Search: 370235
• CPC Class Codes: H04L 41/0813 characterised by the conditions triggering a change of settings
• CPC Class Codes: H04L 41/0895 Configuration of virtualised networks or elements, e.g. virtualised network function or OpenFlow elements
• CPC Class Codes: H04L 41/40 using virtualisation of network functions or resources, e.g. SDN or NFV entities
• CPC Class Codes: H04L 41/5054 Automatic deployment of services triggered by the service manager, e.g. service implementation by automatic configuration of network components
• CPC Class Codes: H04L 43/0817 by checking functioning
• CPC Class Codes: H04L 43/16 Threshold monitoring
• CPC Class Codes: H04L 45/7453 using hashing
• CPC Class Codes: H04L 47/125 by balancing the load, e.g. traffic engineering
• CPC Class Codes: H04L 47/6225 Fixed service order, e.g. Round Robin
• CPC Class Codes: H04L 63/02 for separating internal from external traffic, e.g. firewalls
• CPC Class Codes: H04L 67/1008 based on parameters of servers, e.g. available memory or workload monitoring of computer activity G06F11/30
• CPC Class Codes: H04L 67/1027 Persistence of sessions during load balancing
• CPC Class Codes: H04L 67/1031 Controlling of the operation of servers by a load balancer, e.g. adding or removing servers that serve requests
• CPC Class Codes: H04L 69/22 Parsing or analysis of headers

Abstract

Some embodiments provide an elastic architecture for providing a service in a computing system. To perform a service on the data messages, the service architecture uses a service node (SN) group that includes one primary service node (PSN) and zero or more secondary service nodes (SSNs). The service can be performed on a data message by either the PSN or one of the SSN. However, in addition to performing the service, the PSN also performs a load balancing operation that assesses the load on each service node (i.e., on the PSN or each SSN), and based on this assessment, has the data messages distributed to the service node(s) in its SN group. Based on the assessed load, the PSN in some embodiments also has one or more SSNs added to or removed from its SN group. To add or remove an SSN to or from the service node group, the PSN in some embodiments directs a set of controllers to add (e.g., instantiate or allocate) or remove the SSN to or from the SN group. Also, to assess the load on the service nodes, the PSN in some embodiments receives message load data from the controller set, which collects such data from each service node. In other embodiments, the PSN receives such load data directly from the SSNs.

Claims

1. A system for performing a service on data messages, the system comprising:

a service node (SN) group comprising a primary service node (PSN) and at least one secondary service node (SSN); and

a set of load balancers for directing data message flows to service nodes in the SN group by (i) computing a hash value from the flow'"'"'s header parameter set upon receiving a particular new data message flow, (ii) directing data messages of the particular flow to a particular service node that is associated with a hash value range that contains the computed hash value, and (iii) storing a flow-connection state that identifies the particular service node as the service node of the particular flow for subsequent data messages of the particular flow,

each service node for performing the service on the data messages of the flows that the load balancer (LB) set directs to it,

wherein the PSN is the only SN that provides updated SN group membership data to the LB set, and

wherein based on updated SN group membership data, the LB set modifies how the LB set distributes data message flows to the service nodes of the SN group.

2. The system of claim 1,

wherein the LB set is further for identifying several ranges of hash values that are derivable from header parameter sets of the data message flows and associating each range with a service node;

and

wherein modifying how the LB set distributes data message flows to the service nodes of the SN group comprises modifying the number of hash value ranges.

5. The system of claim 1,

wherein the LB set is further for identifying several ranges of hash values that are derivable from header parameter sets of data message flows and associating each range with a service node;

and

wherein modifying how the LB set distributes data message flows to the service nodes of the SN group comprises modifying the association between the hash value ranges and the service nodes.

6. The system of claim 5, wherein modifying how the LB set distributes data message flows to the service nodes of the SN group further comprises modifying the number of hash value ranges.

9. The system of claim 1, wherein the PSN is further for determining when SN group membership has to be changed, modifying the SN group membership, and after the modification, informing the LB set of the updated SN group membership.

10. The system of claim 9, wherein the PSN modifies the SN group membership by directing a controller set to change the group membership, and upon receiving a confirmation from the controller set of the change, informs the LB set of the updated SN group membership.

11. The system of claim 1, wherein the PSN is further for analyzing statistics regarding the message load data on each service node to determine when the SN group membership has to be changed, modifying the SN group membership, and after the modification, informing the LB set of the updated SN group membership.

12. The system of claim 11, wherein the PSN obtains message load data from a set of SSNs in the SN group.

13. The system of claim 11,

wherein the PSN obtains message load data from a controller set;

wherein the PSN modifies the SN group membership by directing the controller set to change the group membership;

wherein the PSN informs the LB set of the updated SN group membership after receiving confirmation from the controller set that the group membership has been changed.

14. The system of claim 13, wherein the controller set comprises:

a first subset of controllers for collecting message load data from all service nodes, aggregating the message load data and distributing the load data to the PSN; and

a second subset of controllers for modifying the SN group membership by adding or removing service nodes to the group.

15. The system of claim 1, wherein the LB set includes only one load balancer.

16. The system of claim 1, wherein the LB set includes more than one load balancer.

17. The system of claim 1 further comprising:

a plurality of host computing devices,

wherein the SN group comprises a plurality of SSNs,

wherein each service node includes a service virtual machine (SVM) that executes on one host computing device; and

wherein the SVMs of at least two different service nodes execute on two different host computing devices.

3. A system for performing a service on data messages, the system comprising:

a service node (SN) group comprising a primary service node (PSN) and at least one secondary service node (SSN); and

a set of load balancers for directing data message flows to service nodes in the SN group by (i) computing a hash value from the flow'"'"'s header parameter set upon receiving a particular new data message flow, (ii) directing data messages of the particular flow to a particular service node that is associated with a hash value range that contains the computed hash value, and (iii) storing a flow-connection state that identifies the particular service node as the service node of the particular flow for subsequent data messages of the particular flow,

each service node for performing the service on the data messages of the flows that the load balancer (LB) set directs to it,

the PSN further for providing updated SN group membership data to the LB set,

based on updated SN group membership data, the LB set modifying how the LB set distributes data message flows to the service nodes of the SN group by modifying the number of hash value ranges.

4. The system of claim 3, wherein after the LB set modifies its distribution of data message flows to the SN group, the LB set uses the flow-connection state to forward data messages of the particular flow to the particular service node even though, after the updated group membership data, the hash value for the particular flow'"'"'s header parameter set falls within a hash range for another service node.

7. A system for performing a service on data messages, the system comprising:

a service node (SN) group comprising a primary service node (PSN) and at least one secondary service node (SSN), wherein each service node performs the service on the data messages of the flows that are directed to it; and

a set of load balancers (LBs) for directing data message flows to service nodes in the SN group based on current SN group membership data stored by the LB set by (i) computing a hash value from the flow'"'"'s header parameter set upon receiving a particular new data message flow, (ii) directing data messages of the particular flow to a particular service node that is associated with a hash value range that contains the computed hash value, and (iii) storing a flow-connection state that identifies the particular service node as the service node of the particular flow for subsequent data messages of the particular flow,

wherein the PSN further provides updated SN group membership data to the LB set and the SSNs, and

wherein after an SN group membership has been updated but before the LB set updates group membership data that the LB set maintains, a first SSN receives a data message of a first flow that the LB set should have forwarded to a second service node based on the updated SN group membership and forwards the data message to the second service node.

18. The system of claim 7, wherein the LB set distributes the data message flows to the service nodes in a stateless manner.

8. A system for performing a service on data messages, the system comprising:

a service node (SN) group comprising a primary service node (PSN) and at least one secondary service node (SSN), wherein each service node performs the service on the data messages of the flows that are directed to it; and

a set of load balancers (LBs) for directing data message flows to service nodes in the SN group based on current SN group membership data stored by the LB set by (i) computing a hash value from the flow'"'"'s header parameter set upon receiving a particular new data message flow, (ii) directing data messages of the particular flow to a particular service node that is associated with a hash value range that contains the computed hash value, and (iii) storing a flow-connection state that identifies the particular service node as the service node of the particular flow for subsequent data messages of the particular flow,

wherein the PSN and SSN receive group membership updates from a set of controllers that update the group membership,

wherein the PSN further provides updated SN group membership data to the LB set, and

wherein after an SN group membership has been updated but before the LB set updates group membership data that the LB set maintains, a first SSN receives a data message of a first flow that the LB set should have forwarded to a second service node based on the updated SN group membership and forwards the data message to the second service node.

19. The system of claim 8, wherein the LB set distributes the data message flows to the service nodes in a stateless manner.