SCALABLE WORK LOAD MANAGEMENT ON MULTI-CORE COMPUTER SYSTEMS

• US 20120297395A1
• Filed: 04/23/2012
• Published: 11/22/2012
• Est. Priority Date: 08/18/2008
• Status: Abandoned Application

• Current Assignee: Exludus Inc.
• Original Assignee: Exludus Inc.
• Inventors: Marchand, Benoit, Zhou, Xinliang
• Application Number:
  US13/453,099
• Publication Number:
  US 20120297395A1
• Time in Patent Office: Days
• US Class Current: 718/104
• CPC Class Codes: G06F 2209/483 Multiproc
• CPC Class Codes: G06F 2209/485 Resource constraint
• CPC Class Codes: G06F 2209/503 Resource availability
• CPC Class Codes: G06F 9/5027 the resource being a machine, e.g. CPUs, Servers, Terminals

Abstract

A system and method for managing the processing of work units being processed on a computer system having shared resources e.g. multiple processing cores, memory, bandwidth, etc. The system comprises a job scheduler for scheduling access to the shared resources for the work units, and an event trap for capturing resource related allocation events. The event trap is adapted to dynamically adjust the amount of availability associated with each shared resource identified by the resource related allocation event. The allocation event may define a resource release or a resource request. The event trap may increase the amount of availability for allocation events defining a resource release, and decrement the amount of availability for allocation events defining a resource request. The job scheduler allocates resources to the work units using a real time amount of availability of the shared resources in order to maximize a consumption of the shared resources.

Claims

1. A method for managing processing of work units on a computer system having shared resources, the method comprising:

gathering a list of the shared resources in a global resource table, the global resource table indicating an amount of availability for each shared resource;

capturing an allocation event having associated therewith an identification of a shared resource and a value indicating an amount of resource for the identified shared resource;

dynamically adjusting the amount of availability of the identified shared resource based on the value associated with the allocation event for that specific shared resource; and

allocating available resources to the work units which are pending based on a current amount of availability of the shared resources in order to maximize a consumption of the shared resources.

2. The method of claim 1, wherein the dynamically adjusting comprises incrementing the amount of availability of the identified shared resource by the amount of resource indicated by the value for allocation events defining a resource release.

3. The method of claim 1, wherein the dynamically adjusting comprises decrementing the amount of availability of the identified shared resource by the amount of resource indicated by the value for allocation events defining a resource request.

4. The method of claim 1, wherein the capturing comprises implementing an event trap between a system call interface of the computer system and the work units, the event trap for capturing the allocation event.

5. The method of claim 1, further comprising setting, upon initialization of the computer system, the amount of availability for each shared resource to a value that represents 100 percent of capacity of the shared resource.

6. The method as in claim 1, wherein the dynamically adjusting comprises updating the amount of availability of the shared resources, in real time.

7. The method of claim 1, wherein the amount of availability for each shared resource and the amount of resource for the identified shared resource comprises at least one of quantity and time.

8. A non-transitory computer readable medium having recorded thereon one or more programs for execution by a processor for implementing the method of claim 1.

9. A computer system including a processor and a memory having recorded thereon one or more programs for execution by the processor for managing processing of work units using shared resources, the computer system comprising:

a job scheduler for scheduling access to the shared resources for the work units;

an event trap for capturing a resource related allocation event, the event trap being adapted to dynamically adjust an amount of availability associated with each shared resource identified by the resource related allocation event based on a value associated with the resource related allocation event and indicating an amount of resource for the identified shared resource;

wherein the job scheduler allocates resources to the work units using a real time amount of availability of the shared resources in order to maximize a consumption of the shared resources.

10. The system of claim 9, wherein the event trap is provided between a system call interface of the computer system and the work units.

11. The system of claim 9, wherein the event trap is adapted to increment the amount of availability of the identified shared resource by the amount of resource indicated by the value for allocation events defining a resource release.

12. The system of claim 9, wherein the event trap is adapted to decrement the amount of availability of the identified shared resource by the amount of resource indicated by the value for allocation events defining a resource request.

13. The system of claim 9, wherein the shared resources include one or more processors, each having a plurality of processing cores.

14. The system of claim 9, wherein the shared resources include one or more of:

multiple processing core of a heterogeneous nature, shared memory hierarchies with components of heterogeneous access characteristics, shared heterogeneous communications channels, and shared external devices.

15. The system of claim 9, wherein the amount of availability for each shared resource is set to a value that represents 100 percent of capacity of the shared resource, upon initialization.

16. The system of claim 9, wherein processing resources allocated for operation of the event trap are negligible compared to an overall system performance.