Connection pool use of runtime load balancing service performance advisories

• US 7,937,493 B2
• Filed: 06/27/2005
• Issued: 05/03/2011
• Est. Priority Date: 08/14/2003
• Status: Active Grant

• Current Assignee: Oracle International Corporation (Oracle Corporation)
• Original Assignee: Oracle International Corporation (Oracle Corporation)
• Inventors: Colrain, Carol L., Surber, Douglas N., Irudayaraj, Rajkumar A.
• Primary Examiner(s): NGUYEN, THANH T
• Application Number:
  US11/168,968
• Publication Number:
  US 20050262183A1
• Time in Patent Office: 2,136 Days
• Field of Search: 709/224, 709/226, 709/244, 370/408, 379/201.2
• US Class Current: 709/244
• CPC Class Codes: G06F 2209/508 Monitor
• CPC Class Codes: G06F 9/505 considering the load
• CPC Class Codes: G06F 9/5083 Techniques for rebalancing the load in a distributed system

Abstract

Runtime connection load balancing of work across connections to a clustered computing system involves the routing of requests for a service, based on the current operational performance of each of the instances that offer the service. A connection is selected from an identified connection pool, to connect to an instance that provides the service for routing a work request. The operational performance of the instances may be represented by performance information that characterizes the response time and/or the throughput of the service that is provided by a particular instance on a respective node of the system, and is relative to other instances that offer the same service.

Claims

1. A computer-implemented method for routing work requests to a computing system that comprises a plurality of computing nodes that each hosts a server instance that provides a service that performs work, the method comprising:

receiving a request for a connection to said computing system for requesting said service;

identifying a pool of connections that is associated with said service, wherein each connection in said pool is a connection to, and recognized by, a particular server instance;

wherein a first connection associated with said service is a connection to and recognized by a first server instance and a second connection associated with said service is a connection to and recognized by a second server instance, wherein the first connection is not the same as the second connection, and the first server instance is not the same as the second server instance;

wherein said each connection is one of a plurality of connections that simultaneously existed in said pool prior to receiving said request for a connection; and

selecting, from said pool, a connection that is associated with said service, to route a request for performance of work associated with said service;

wherein said selecting is based on substantially current performance information about server instances that offer said service that is associated with said selected connection;

wherein the performance information for each of said instances that offer said service includes a value representing a percentage of work to route to said instance;

wherein the method is performed by one or more computing devices.

2. The method of claim 1, wherein selecting a connection to route the request comprises:

(a) computing the sum of said percentages for valid instances providing said service;

(b) computing a pseudo-random number that is greater than or equal to zero and less than said sum of said percentages;

(c) set a variable value to zero;

(d) creating a set of candidate instances from the valid instances;

(e) repeating the following steps in order until an instance is selected;

(f) randomly choosing a particular instance from the set of candidate instances for consideration;

(g) summing said percentage for said particular instance and said variable value; and

(h) if said pseudo-random number is less than said variable value, then selecting said

particular instance, otherwise removing from consideration said particular instance from

said candidate instances.

3. The method of claim 2, further comprising:

retrieving, from said pool, said selected connection to route the request.

4. The method of claim 2, further comprising:

based on a particular ordering, ordering the set of candidate instances;

wherein a next candidate instance to be chosen from the ordered set of candidate instances is the first candidate instance, based on said particular ordering, remaining in the ordered set of candidate instances;

wherein the step of choosing the particular instance comprises choosing the next candidate instance.

5. The method of claim 1, further comprising:

selecting a connection, from said pool, to retire from said pool.

6. The method of claim 5, wherein selecting a connection to retire includes:

selecting an instance from which to retire said connection by computing, for each instance that offers said service,

a first ratio of the number of requests for a connection to said instance over a total number of requests for said service;

a second ratio of the number of connections to said instance over a total number of connections in said pool; and

selecting said instance from which to retire said connection based on the difference between said second ratio and said first ratio.

7. The method of claim 1,

wherein said performance information comprises a characterization of any one or more of (a) a response time of said service and (b) a throughput of said service, associated with a respective instance providing said service on a node of said computing system.

8. A non-transitory machine-readable storage medium storing one or more sequences of instructions which, when executed by one or more processors, causes the one or more processors to perform:

receiving a request for a connection to said computing system for requesting said service;

receiving a request for a connection to said computing system for requesting said service;

identifying a pool of connections that is associated with said service, wherein each connection in said pool is a connection to, and recognized by, a particular server instance;

wherein a first connection associated with said service is a connection to and recognized by a first server instance and a second connection associated with said service is a connection to and recognized by a second server instance, wherein the first connection is not the same as the second connection, and the first server instance is not the same as the second server instance;

wherein said each connection is one of a plurality of connections that simultaneously existed in said pool prior to receiving said request for a connection; and

selecting, from said pool, a connection that is associated with said service, to route a request for performance of work associated with said service;

wherein said selecting is based on substantially current performance information about server instances that offer said service that is associated with said selected connection;

wherein the performance information for each of said instances that offer said service includes a value representing a percentage of work to route to said instance.

9. The non-transitory machine-readable storage medium of claim 8, wherein selecting a connection to route the request comprises:

(a) computing the sum of said percentages for valid instances providing said service;

(b) computing a pseudo-random number that is greater than or equal to zero and less than said sum of said percentages;

(c) set a variable value to zero;

(d) creating a set of candidate instances from the valid instances;

(e) repeating the following steps in order until an instance is selected;

(f) randomly choosing a particular instance from the set of candidate instances for consideration;

(g) summing said percentage for said particular instance and said variable value; and

(h) if said pseudo-random number is less than said variable value, then selecting said particular instance, otherwise removing from consideration said particular instance from said candidate instances.

10. The non-transitory machine-readable storage medium of claim 9, further comprising:

retrieving, from said pool, said selected connection to route the request.

11. The non-transitory machine-readable storage medium of claim 9, further comprising:

based on a particular ordering, ordering the set of candidate instances;

wherein a next candidate instance to be chosen from the ordered set of candidate instances is the first candidate instance, based on said particular ordering, remaining in the ordered set of candidate instances;

wherein the step of choosing the particular instance comprises choosing the next candidate instance.

12. The non-transitory machine-readable storage medium of claim 8, further comprising:

selecting a connection, from said pool, to retire from said pool.

13. The non-transitory machine-readable storage medium of claim 12, wherein selecting a connection to retire includes:

selecting an instance from which to retire said connection by computing, for each instance that offers said service,

a first ratio of the number of requests for a connection to said instance over a total number of requests for said service;

a second ratio of the number of connections to said instance over a total number of connections in said pool; and

selecting said instance from which to retire said connection based on the difference between said second ratio and said first ratio.

14. The non-transitory machine-readable storage medium of claim 8, wherein said performance information comprises a characterization of any one or more of (a) a response time of said service and (b) a throughput of said service, associated with a respective instance providing said service on a node of said computing system.

15. A computer-implemented method for routing work requests to a computing system that comprises a plurality of computing nodes that each hosts a server instance that provides a service that performs work, the method comprising:

identifying a pool of connections that is associated with said service, wherein each connection in said pool is a connection to, and recognized by, a particular server instance;

wherein a first connection associated with said service is a connection to and recognized by a first server instance and a second connection associated with said service is a connection to and recognized by a second server instance, wherein the first connection is not the same as the second connection, and the first server instance is not the same as the second server instance;

wherein said each connection is one of a plurality of connections that simultaneously existed in said pool prior to receiving a request for a connection; and

selecting, from said pool, a connection that is associated with said service, to route a request for performance of work associated with said service;

wherein said selecting is based on substantially current performance information about server instances that offer said service that is associated with said selected connection;

selecting a connection, from said pool, to retire from said pool;

wherein the method is performed by one or more computing devices.

16. The method of claim 15, wherein selecting a connection to retire includes:

selecting an instance from which to retire said connection by computing, for each instance that offers said service,

a first ratio of the number of requests for a connection to said instance over a total number of requests for said service;

a second ratio of the number of connections to said instance over a total number of connections in said pool; and

selecting said instance from which to retire said connection based on the difference between said second ratio and said first ratio.

17. A non-transitory machine-readable storage medium storing one or more sequences of instructions which, when executed by one or more processors, causes the one or more processors to perform:

identifying a pool of connections that is associated with said service, wherein each connection in said pool is a connection to, and recognized by, a particular server instance;

wherein a first connection associated with said service is a connection to and recognized by a first server instance and a second connection associated with said service is a connection to and recognized by a second server instance, wherein the first connection is not the same as the second connection, and the first server instance is not the same as the second server instance;

wherein said each connection is one of a plurality of connections that simultaneously existed in said pool prior to receiving a request for a connection; and

selecting, from said pool, a connection that is associated with said service, to route a request for performance of work associated with said service;

wherein said selecting is based on substantially current performance information about server instances that offer said service that is associated with said selected connection;

selecting a connection, from said pool, to retire from said pool.

18. The non-transitory machine-readable storage medium of claim 17, wherein selecting a connection to retire includes:

selecting an instance from which to retire said connection by computing, for each instance that offers said service,

a first ratio of the number of requests for a connection to said instance over a total number of requests for said service;

a second ratio of the number of connections to said instance over a total number of connections in said pool; and

selecting said instance from which to retire said connection based on the difference between said second ratio and said first ratio.

19. A computer-implemented method for routing work requests to a computing system that comprises a plurality of computing nodes that each hosts a server instance that provides a service that performs work, the method comprising:

receiving a request for a connection to said computing system for requesting said service;

identifying a pool of connections that is associated with said service, wherein each connection in said pool is a connection to, and recognized by, a particular server instance;

wherein a first connection associated with said service is a connection to and recognized by a first server instance and a second connection associated with said service is a connection to and recognized by a second server instance, wherein the first connection is not the same as the second connection, and the first server instance is not the same as the second server instance;

wherein said each connection is one of a plurality of connections that simultaneously existed in said pool prior to receiving said request for a connection; and

selecting, from said pool, a connection that is associated with said service, to route a request for performance of work associated with said service;

wherein said selecting is based on substantially current performance information about server instances that offer said service that is associated with said selected connection;

selecting a connection, from said pool, to retire from said pool; and

wherein the method is performed by one or more computing devices.

20. A non-transitory machine-readable storage medium storing one or more sequences of instructions which, when executed by one or more processors, causes the one or more processors to perform:

receiving a request for a connection to a computing system for requesting a service;

identifying a pool of connections that is associated with said service, wherein each connection in said pool is a connection to, and recognized by, a particular server instance;

wherein a first connection associated with said service is a connection to and recognized by a first server instance and a second connection associated with said service is a connection to and recognized by a second server instance, wherein the first connection is not the same as the second connection, and the first server instance is not the same as the second server instance;

wherein said each connection is one of a plurality of connections that simultaneously existed in said pool prior to receiving said request for a connection; and

selecting, from said pool, a connection that is associated with said service, to route a request for performance of work associated with said service;

wherein said selecting is based on substantially current performance information about server instances that offer said service that is associated with said selected connection;

selecting a connection, from said pool, to retire from said pool.