System and method for application server with self-tuned threading model

• US 7,657,892 B2
• Filed: 05/19/2005
• Issued: 02/02/2010
• Est. Priority Date: 05/20/2004
• Status: Active Grant

• Current Assignee: Oracle International Corporation (Oracle Corporation)
• Original Assignee: BEA Systems Incorporated (Oracle Corporation)
• Inventors: Langen, Anno, Revanuru, Naresh
• Primary Examiner(s): Zhen; Li B
• Application Number:
  US11/132,601
• Publication Number:
  US 20050262507A1
• Time in Patent Office: 1,720 Days
• Field of Search: 719/314, 718/100, 718/103, 718/104, 718/102
• US Class Current: 718/102
• CPC Class Codes: G06F 2209/5011 Pool
• CPC Class Codes: G06F 2209/5018 Thread allocation
• CPC Class Codes: G06F 2209/5021 Priority
• CPC Class Codes: G06F 2209/504 Resource capping
• CPC Class Codes: G06F 9/4881 Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues
• CPC Class Codes: G06F 9/5038 considering the execution order of a plurality of tasks, e.g. taking priority or time dependency constraints into consideration scheduling strategies G06F9/4881 and subgroups
• CPC Class Codes: Y02D 10/00 Energy efficient computing, e.g. low power processors, power management or thermal management

Abstract

A system and method for application server with self-tuned threading model. A server queue is used as a priority scheme, including a number of entries therein associated with received requests, and which allows the entries to have expressed priorities or share values, instead of simply the number of threads. Flexibility is retained to express the threads in raw numbers whenever this might be desired, or to express constraints on the number of threads to be made available.

Claims

1. A system for self-tuned threading model in an application server, comprising:

a server, including one or more processors, which receives requests from clients or other servers for execution by service classes at the server;

a plurality of threads operating at the server for use in processing the requests;

a priority queue operating at the server for storing the requests until the requests can be processed by the threads;

a configuration file, which defines a plurality of service classes executing at the server to which each of the requests can be directed, wherein the configuration file includes numeric share values associated with the plurality of service classes, including a share value for each service class, and wherein each of the share values represents a proportional entitlement to thread usage for its associated service class as compared to other service classes having other share values;

a software process that operates on the server and maintains the priority queue for allocating threads to requests, according to the service class to which the requests are directed and the proportional entitlement priority to thread usage for each service class, including

determining that service class to which a received request is directed,

determining the share value associated with that service class, and

placing that request in the priority queue in proportion to the share value for each request'"'"'s service class, so that the threads are provided with requests for each service class in a ratio matching those service classes proportional entitlement; and

wherein one or more numeric share values associated with the plurality of service classes and defined in the configuration file can be modified, to reflect a different proportional entitlement, and wherein the software process automatically modifies the placement of subsequent received requests in the priority queue, to reflect the different proportional entitlement.

2. The system of claim 1 wherein said server includes multiple processors.

3. The system of claim 1 wherein the entries are queued in the priority queue according to pro-rata time values.

4. The system of claim 3 wherein the entries are queued according to the calculated thread usage time already used in comparison to the time values.

5. The system of claim 1 wherein the entries may be queued according to additional configured constraints.

6. The system of claim 5 wherein the constraints are determined by queuing the entry in an additional constraints queue.

7. The system of claim 1 wherein the placing of the request in the priority queue and presenting to the threads the request with a priority in proportion to the share value further comprises

lowering the priority by an amount proportional to amount of threads and thread time already used by the client making the request, and raising or lowering by an amount related to the response time goal.

8. A method for self-tuned threading model in an application server, comprising the steps of:

configuring the server to share thread resources among entities according to share values;

receiving requests at the server from clients or other servers for execution by service classes at the server;

determining a plurality of service classes to which each requests can be directed, wherein a configuration file includes numeric share values associated with the plurality of service classes, including a share value for each service class, and wherein each of the share values represents a proportional entitlement to thread usage for its associated service class as compared to other service classes having other share values;

queuing requests in a priority queue in proportion to the share value for each requests'"'"' service class;

maintaining the priority queue for allocating threads to requests, according to the service class to which the requests are directed and the proportional entitlement priority to thread usage for each service class; and

taking the requests from the priority queue and presenting the requests to the threads for processing by a processor, wherein for each request the method further comprises

determining that service class to which a received request is directed,

determining the share value associated with that service class, and

placing that request in the priority queue in proportion to the share value for each request'"'"'s service class, so that the threads are provided with requests for each service class in a ratio matching those service classes proportional entitlement; and

wherein one or more numeric share values associated with the plurality of service classes and defined in the configuration file can be modified, to reflect a different proportional entitlement, and wherein the software process automatically modifies the placement of subsequent received requests in the priority queue, to reflect the different proportional entitlement.

9. The method of claim 8 wherein said sewer includes multiple processors.

10. The method of claim 8 wherein the entries are queued in the priority queue according to pro-rata time values.

11. The method of claim 10 wherein the entries are queued according to the calculated thread usage time already used in comparison to the time values.

12. The method of claim 8 wherein the entries may be queued according to additional configured constraints.

13. The method of claim 12 wherein the constraints are determined by queuing the entry in an additional constraints queue.

14. The method of claim 8 wherein the placing of the request in the priority queue and presenting to the threads the request with a priority in proportion to the share value further comprises

lowering the priority by an amount proportional to amount of threads and thread time already used by the client making the request, and raising or lowering by an amount related to the response time goal.

15. A computer readable storage medium, including instructions thereon which when executed cause the computer to perform the steps of:

configuring the server to share thread resources among entities according to share values;

receiving requests at the sewer from clients or other servers for execution by service classes at the server;

determining a plurality of service classes to which each requests can be directed, wherein a configuration file includes numeric share values associated with the plurality of service classes, including a share value for each service class, and wherein each of the share values represents a proportional entitlement to thread usage for its associated service class as compared to other service classes having other share values;

queuing requests in a priority queue in proportion to the share value for each requests'"'"' service class;

maintaining the priority queue for allocating threads to requests, according to the service class to which the requests are directed and the proportional entitlement priority to thread usage for each service class; and

taking the requests from the priority queue and presenting the requests to the threads for processing by a processor, wherein for each request the method further comprises

determining that service class to which a received request is directed,

determining the share value associated with that service class, and

placing that request in the priority queue in proportion to the share value for each request'"'"'s service class, so that the threads are provided with requests for each service class in a ratio matching those service classes proportional entitlement; and

wherein one or more numeric share values associated with the plurality of service classes and defined in the configuration file can be modified, to reflect a different proportional entitlement, and wherein the software process automatically modifies the placement of subsequent received requests in the priority queue, to reflect the different proportional entitlement.

16. The computer readable storage medium of claim 15 wherein said server includes multiple processors.

17. The computer readable storage medium of claim 15 wherein the entries are queued in the priority queue according to pro-rata time values.

18. The computer readable storage medium of claim 17 wherein the entries are queued according to the calculated thread usage time already used in comparison to the time values.

19. The computer readable storage medium of claim 15 wherein the entries may be queued according to additional configured constraints.

20. The computer readable storage medium of claim 19 wherein the constraints are determined by queuing the entry in an additional constraints queue.

21. The computer readable storage medium of claim 15 wherein the placing of the request in the priority queue and presenting to the threads the request with a priority in proportion to the share value further comprises

lowering the priority by an amount proportional to amount of threads and thread time already used by the client making the request, and raising or lowering by an amount related to the response time goal.