SYSTEM AND METHOD FOR NON-DISRUPTIVE MANAGEMENT OF SERVERS IN A NETWORK ENVIRONMENT

• US 20130155902A1
• Filed: 12/16/2011
• Published: 06/20/2013
• Est. Priority Date: 12/16/2011
• Status: Active Grant

• Current Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Original Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Inventors: Feng, Chao, Sharma, Samar, Desai, Ronak, Qu, Diheng
• Granted Patent: US 9,088,584 B2
• Time in Patent Office: Days
• US Class Current: 370/255
• CPC Class Codes: G06F 11/16 Error detection or correction of the data by redundancy in hardware
• CPC Class Codes: G06F 11/2007 using redundant communication media
• CPC Class Codes: G06F 11/2028 eliminating a faulty processor or activating a spare
• CPC Class Codes: G06F 11/203 using migration
• CPC Class Codes: G06F 11/2035 without idle spare hardware
• CPC Class Codes: G06F 11/2038 with a single idle spare processing component
• 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

Abstract

An example method includes disengaging a target node from a cluster, where the disengaging comprises: selecting an inheritor; migrating flows from the target node to the inheritor; informing a migration manager that the target node is disengaged from the cluster; and broadcasting to peer nodes of the target node that the target node is replaced by the inheritor. In particular implementations of the present disclosure, the cluster can include a first layer of a network topology including a forwarding engine that implements hash-based packet forwarding; a second layer of the network topology comprising the target node and the inheritor, where the target node and the inheritor implement flow-based packet forwarding; and a third layer including service nodes configured for packet processing in a network.

Claims

1. A method comprising:

disengaging a target node from a cluster, wherein the disengaging comprises;

selecting an inheritor;

migrating flows from the target node to the inheritor;

informing a migration manager that the target node is disengaged from the cluster; and

broadcasting to peer nodes of the target node that the target node is replaced by the inheritor.

2. The method of claim 1, wherein the cluster comprises:

a first layer of a network topology including a forwarding engine that implements hash-based packet forwarding; and

a second layer of the network topology comprising the target node and the inheritor, wherein the target node and the inheritor implement flow-based packet forwarding.

3. The method of claim 2, wherein the cluster further comprises:

a third layer including service nodes configured for packet processing in a network.

4. The method of claim 1, wherein the inheritor comprises a peer node of the target node having a least number of flows within a subset of nodes of the cluster.

5. The method of claim 1, wherein the inheritor comprises a new peer node in the cluster.

6. The method of claim 1, wherein selecting the inheritor comprises:

receiving flow handling statistics of peer nodes from the migration manager; and

distributing the flows among the peer nodes based on the flow handling statistics so that a total number of flows being handled by the peer nodes are evenly distributed.

7. The method of claim 1, wherein the target node comprises a flow-packet distributor.

8. The method of claim 1, wherein the migrating comprises duplicating a flow table to the inheritor.

9. The method of claim 8, further comprising:

purging the flow table;

identifying that the flow table is empty and no data packet is pending in a receive/transmit (RX/TX) queue of the target node; and

setting a status of the target node to indicate that the target node is disengaged from the cluster.

10. The method of claim 1, further comprising:

redirecting a packet from the target node to the inheritor.

11. The method of claim 1, further comprising:

providing instructions for a removal of the target node;

identifying a confirmation that the target node is removed; and

changing a hash scope in a hashing algorithm to remove the target node.

12. An apparatus, comprising:

a memory configured to store instructions;

a processor operable to execute the instructions;

an inheritor selector module;

a flow handler; and

a broadcast module, wherein the apparatus is configured for;

disengaging a target node from a cluster, wherein the disengaging comprises;

selecting an inheritor;

migrating flows from the target node to the inheritor;

informing a migration manager that the target node is disengaged from the cluster; and

broadcasting to peer nodes of the target node that the target node is replaced by the inheritor.

13. The apparatus of claim 12, wherein the cluster comprises:

a first layer of a network topology including a forwarding engine that implements hash-based packet forwarding; and

a second layer of the network topology comprising the target node and the inheritor, wherein the target node and the inheritor implement flow-based packet forwarding.

14. The apparatus of claim 12, wherein the migrating comprises providing a flow table to the inheritor.

15. The apparatus of claim 14, wherein the apparatus is further configured for:

purging the flow table;

identifying that the flow table is empty and no data packet is pending in a RX/TX queue of the target node; and

setting a status of the target node to indicate that the target node is disengaged from the cluster.

16. Logic encoded in non-transitory media that includes code for execution and when executed by a processor is operable to perform operations comprising:

disengaging a target node from a cluster, wherein the disengaging comprises;

selecting an inheritor;

migrating flows from the target node to the inheritor;

informing a migration manager that the target node is disengaged from the cluster; and

broadcasting to peer nodes of the target node that the target node is replaced by the inheritor.

17. The logic of claim 16, wherein the cluster comprises:

a first layer of a network topology including a forwarding engine that implements hash-based packet forwarding; and

a second layer of the network topology comprising the target node and the inheritor, wherein the target node and the inheritor implement flow-based packet forwarding.

18. The logic of claim 16, wherein the inheritor comprises a peer node of the target node having a least number of flows within a subset of nodes of the cluster.

19. The logic of claim 16, wherein the migrating comprises duplicating a flow table to be provided to the inheritor.

20. The logic of claim 19, the operations further comprising:

purging the flow table;

identifying that the flow table is empty and no data packet is pending in a RX/TX queue of the target node; and

setting a status of the target node to indicate that the target node is disengaged from the cluster.