Load switch command including identification of source server cluster and target server cluster

Patent metadata

Field	Value
Publication number	US-2016112503-A1
Application number	US-201414895186-A
Country	US
Kind code	A1
Filing date	Jun 6, 2014
Priority date	Jun 9, 2013
Publication date	Apr 21, 2016
Grant date	—

Abstract

A software defined network (SDN) controller sends a load switch command to a network device. The load switch command includes identification of a source server cluster and identification of a target server cluster.

First claim

1 . A method for load balancing across clusters, including:
obtaining, by a software defined network (SDN) controller, load information which is reported by a source server cluster; determining based on the load information that the source server cluster is in an overload state; in response to determining that the source server cluster is in an overload state, updating the state of the source server cluster in a cross-cluster table of the SDN controller; searching the cross-cluster table for a target server cluster which is in an idle state and which is located in a server cluster group which the source server cluster belongs to; sending a load switch command to a predetermined network device which is located in a front path of the source server cluster, wherein the load switch command includes identification of the source server cluster and identification of a target server cluster.

2 . The method according to claim 1 , further comprising:
periodically and synchronously transmitting a current session list from the SDN controller to the predetermined network device.

3 . The method according to claim 1 , further comprising:
sending a load switch-back command to the predetermined network device in response to a determination that the source server cluster has exited from the overload state, wherein the load switch-back command includes identification of the source server cluster.

4 . The method according to claim 1 , comprising determining that the source service cluster is in the overload state when a load of the first service cluster exceeds a first threshold and determining that the source service cluster has exited the overload state when a load of the source server cluster has dropped below a second threshold;
wherein the first threshold is higher than the second threshold.

5 . A method for transferring access across clusters, comprising:
receiving, by a network device, a load switch command from a software defined network (SDN) controller, wherein the load switch command includes identification of a source server cluster and identification of a target server cluster, wherein the source server cluster and the target server cluster belong to a server cluster group; and performing a switch processing mode for the source server cluster; wherein the process of performing a switch processing mode comprises: receiving a packet for accessing the source server cluster; determining whether the packet belongs to a session which has already been established for the source server cluster, and if the packet belongs to a session which has already been established for the source server cluster, forwarding the packet to the source server cluster, otherwise, acting as proxy for an external host to access the target server cluster.

6 . A method according to claim 5 , further comprising:
periodically receiving a session list of the source server cluster from the SDN controller.

7 . A method according to claim 6 , further comprising:
creating a record in a proxy session table when the external host accesses the target server cluster by proxy; receiving a load switch-back command from the SDN controller, performing a switch-back processing mode for a first server cluster based on an identification of the first server cluster in the load switch-back command; wherein the process of performing the switch-back processing mode comprises: receiving a packet from an external host to request access the source service cluster; determining if the packet matches with a record in the proxy session table; and if the packet matches with a record in the proxy session table, then acting as proxy for the external host to access the target server cluster.

8 . A non-transitory computer readable storage medium on which is stored machine readable instructions that when executed by a processor cause the processor to:
obtain load information reported by a source server cluster; update the state of the source server cluster in a cross-cluster table in accordance with the load information; in response to determining that the source server cluster is in an overload state, search the cross-cluster table for a target server cluster that is in an idle state and which is located in a same server cluster group as the source server cluster, and send a load switch command to a predetermined network device which is located in a front path of the source server cluster, wherein the load switch command includes identification of the source server cluster and identification of a target server cluster.

9 . The non-transitory computer readable storage medium according to claim 8 , wherein the machine readable instructions are further to cause the processor to:
periodically and synchronously transmit a current session list to the predetermined network device in synchronicity in which the current session list is send from the source server cluster to the processor when the processor determines that the source server cluster is in the overload state.

10 . The non-transitory computer readable storage medium according to claim 8 , wherein the machine readable instructions are further to cause the processor to:
send a load switch-back command to the predetermined network device when the source server cluster exits from the overload state, wherein the load switch-back command includes identification of the source server cluster.

11 . The non-transitory computer readable storage medium according to claim 8 ,
wherein the machine readable instructions are further to cause the processor to: determine that the source service cluster is in the overload state when a load of the first service cluster exceeds a first threshold; and determine that the source service cluster has exited the overload state when a load of the source server cluster has dropped below a second threshold; wherein the first threshold is higher than the second threshold.

12 . A non-transitory computer readable storage medium on which is stored machine readable instructions that when executed by a processor cause the processor to:
receive a load switch command from a software defined networks (SDN) controller, wherein the load switch command has identification of a source server cluster and identification of a target server cluster, wherein the source server cluster and the target server cluster belong to a server cluster group; perform a switch processing mode for the source server cluster; receive a packet for access the source server cluster; determine whether the packet is a session which has already been established for the source server cluster, and if the packet is a session which has already been established for the source server cluster, forward the packet to the source server cluster, otherwise, act as proxy for an external host to access the target server cluster.

13 . The non-transitory computer readable storage medium according to claim 12 , wherein the machine readable instructions are further to cause the processor to:
periodically receive session list of the source server cluster from the SDN controller.

14 . The non-transitory computer readable storage medium according to claim 13 , wherein the machine readable instructions are further to cause the processor to:
create a record in a proxy session table when the external host accesses the target server cluster by proxy; receive a load switch-back command from the SDN controller, perform a switch-back mode for a first server cluster based on an identification of the first server cluster in the load switch-back instruction; and determine whether a packet matches with a record in the proxy session table when the packe

Assignees

• Hangzhou H3C Tech Co Ltd
• Hangzhou H3C Tech Co Ltd

Inventors

• Yang Qi

Classifications

• H04L67/1027 Primary

  Persistence of sessions during load balancing · CPC title

• H04L67/1008 Primary

  based on parameters of servers, e.g. available memory or workload (monitoring of computer activity G06F11/30) · CPC title

• H04L12/6418 Primary

  Hybrid transport · CPC title

• H04L47/125

  by balancing the load, e.g. traffic engineering · CPC title

• H04L49/50

  Overload detection or protection within a single switching element · CPC title