Load balancing of distributed services

• US 10,044,797 B2
• Filed: 08/23/2017
• Issued: 08/07/2018
• Est. Priority Date: 03/26/2014
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Guerin, Xavier R.
• Primary Examiner(s): Nguyen, Dustin
• Assistant Examiner(s): Guzman, Javier O
• Application Number:
  US15/684,209
• Publication Number:
  US 20170353538A1
• Time in Patent Office: 349 Days
• CPC Class Codes: H04L 43/08 Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
• CPC Class Codes: H04L 67/1004 Server selection for load balancing
• CPC Class Codes: H04L 67/1008 based on parameters of servers, e.g. available memory or workload monitoring of computer activity G06F11/30

Abstract

Various embodiments load balance service requests across one or more servers. In one embodiment, a service requestor directly accesses a shared metrics array stored in at least one server of a plurality of servers in a service cluster. Each of these servers includes one or more services. The shared metrics array is accessible by each of the plurality of servers, and includes a set of metrics for each of the plurality of servers. A determination is made based on the set of metrics associated with at least one server in the plurality of servers if a service request is to be sent to the at least one server. The service request is sent to the at least one server based on determining that the service request is to be sent to the at least one sever.

Claims

1. A method, with an information processing system, for load balancing service requests across one or more servers, the method comprising:

accessing, by the information processing system in response to obtaining a service request, a shared metrics array stored in memory of at least one server of a plurality of servers in a service cluster, wherein the accessing comprises performing one or more remote direct memory access operations on a memory of the at least one server comprising the shared metrics array,

where the information processing system comprises the service request and is external to the service cluster, and where each server in the plurality of servers comprises at least one service for satisfying the service request, the shared metrics array being accessible at the server by each of the plurality of servers, wherein the shared metrics array comprises a set of metrics for each of the plurality of servers, and wherein each of the plurality of servers comprises one or more services, the shared metrics array being directly accessed in the memory of the server by the information processing system without involving a processor of the at least one server;

determining, based on the set of metrics associated with at least one server in the plurality of servers, if the service request is to be sent to the at least one server; and

sending the service request to the at least one server based on determining that the service request is to be sent to the at least one sever.

2. The method of claim 1, further comprising:

selecting the at least one server based on one or more selection policies.

3. The method of claim 1, further comprising:

based determining that the service request is not to be sent to the at least one server;

selecting at least one additional server from the plurality of servers; and

sending the service request to the at least one additional server based on the set of metrics associated with the at least one additional server.

4. The method of claim 1, wherein the shared metrics array comprises a plurality of slots, wherein each slot in the plurality of slots is associated with one of the plurality of servers and comprises the set of metrics for the one of the plurality of servers.

5. The method of claim 4, a given slot in the plurality of slots is updatable only by its associated server in the plurality of servers.

6. The method of claim 4, wherein each slot in the plurality of slots further comprises at least one of:

a timestamp associated with a most recent update to set of metrics stored within the slot;

an identifier associated with the server associated with the slot; and

a set of connection information for the server associated with the slot.

7. An information processing system for load balancing service requests across one or more servers, the information processing system comprising:

a processor;

a memory communicatively coupled to the processor; and

a load balancer communicatively coupled to the processor and the memory, the load balancer configured to perform a method comprising;

accessing, in response to obtaining a service request, a shared metrics array stored in memory of at least one server of a plurality of servers in a service cluster, wherein the accessing comprises performing one or more remote direct memory access operations on a memory of the at least one server comprising the shared metrics array,

where the information processing system is external to the service cluster and comprises service request, and where each server in the plurality of servers comprises at least one service for satisfying the service request, the shared metrics array being accessible at the server by each of the plurality of servers, wherein the shared metrics array comprises a set of metrics for each of the plurality of servers, and wherein each of the plurality of servers comprises one or more services, the shared metrics array being directly accessed in the memory of the server by the load balancer without involving a processor of the at least one server;

determining, based on the set of metrics associated with at least one server in the plurality of servers, if the service request is to be sent to the at least one server; and

sending the service request to the at least one server based on determining that the service request is to be sent to the at least one sever.

8. The information processing system of claim 7, wherein the shared metrics array comprises a plurality of slots, wherein each slot in the plurality of slots is associated with one of the plurality of servers and comprises the set of metrics for the one of the plurality of servers.

9. The information processing system of claim 8, wherein each slot in the plurality of slots further comprises at least one of:

a timestamp associated with a most recent update to set of metrics stored within the slot;

an identifier associated with the server associated with the slot; and

a set of connection information for the server associated with the slot.

10. A computer program storage product for load balancing service requests across one or more servers, the computer program storage product comprising a non-transitory storage medium readable storing instructions for execution by an information processing system for performing a method comprising:

accessing, by the information processing system in response to obtaining a service request, a shared metrics array stored in memory of at least one server of a plurality of servers in a service cluster, wherein the accessing comprises performing one or more remote direct memory access operations on a memory of the at least one server comprising the shared metrics array,

where the information processing system comprises the service request and is external to the service cluster, and where each server in the plurality of servers comprises at least one service for satisfying the service request, the shared metrics array being accessible at the server by each of the plurality of servers, wherein the shared metrics array comprises a set of metrics for each of the plurality of servers, and wherein each of the plurality of servers comprises one or more services, the shared metrics array being directly accessed in the memory of the server by the information processing system without involving a processor of the at least one server;

determining, based on the set of metrics associated with at least one server in the plurality of servers, if the service request is to be sent to the at least one server; and

sending the service request to the at least one server based on determining that the service request is to be sent to the at least one sever.

11. The computer program storage product of claim 10, wherein the method further comprises:

based determining that the service request is not to be sent to the at least one server;

selecting at least one additional server from the plurality of servers; and

sending the service request to the at least one additional server based on the set of metrics associated with the at least one additional server.

12. The computer program storage product of claim 10, wherein the shared metrics array comprises a plurality of slots, wherein each slot in the plurality of slots is associated with one of the plurality of servers and comprises the set of metrics for the one of the plurality of servers.

13. The computer program storage product of claim 12, a given slot in the plurality of slots is updatable only by its associated server in the plurality of servers.

14. The computer program storage product of claim 12, wherein each slot in the plurality of slots further comprises at least one of:

a timestamp associated with a most recent update to set of metrics stored within the slot;

an identifier associated with the server associated with the slot; and

a set of connection information for the server associated with the slot.