Dynamic Feedback Load Balancing

• US 20120001925A1
• Filed: 06/30/2011
• Published: 01/05/2012
• Est. Priority Date: 06/30/2010
• Status: Abandoned Application

• Current Assignee: ATI Technologies ULC (Advanced Micro Devices, Inc.)
• Original Assignee: ATI Technologies ULC (Advanced Micro Devices, Inc.)
• Inventors: ANDONIEH, Joseph, RAHMAN, Arshad
• Application Number:
  US13/173,995
• Publication Number:
  US 20120001925A1
• Time in Patent Office: Days
• US Class Current: 345/502
• CPC Class Codes: A63F 13/52 involving aspects of the displayed game scene
• CPC Class Codes: A63F 2300/203 Image generating hardware
• CPC Class Codes: G06F 15/16 Combinations of two or more digital computers each having at least an arithmetic unit, a program unit and a register, e.g. for a simultaneous processing of several programs coordinating program control therefor G06F9/52; in regulating and control system G05B
• CPC Class Codes: G06F 2209/508 Monitor
• CPC Class Codes: G06F 9/5083 Techniques for rebalancing the load in a distributed system
• CPC Class Codes: G06T 1/20 Processor architectures; Processor configuration, e.g. pipelining
• CPC Class Codes: G06T 1/60 Memory management
• CPC Class Codes: G06T 15/005 General purpose rendering architectures

Abstract

A method for rendering a scene across N number of processors is provided. The method includes evaluating performance statistics for each of the processors and establishing load rendering boundaries for each of the processors, the boundaries defining a respective portion of the scene. The method also includes dynamically adjusting the boundaries based upon the establishing and the evaluating.

Claims

1. A method for rendering a scene across N number of processors, comprising:

evaluating performance statistics for each of the processors;

establishing load rendering boundaries for each of the processors, the boundaries defining a respective portion of the scene; and

dynamically adjusting the boundaries based upon the establishing and the evaluating.

2. The method of claim 1, wherein the processors are graphics processing units (GPUs).

3. The method of claim 2, wherein the evaluating includes calculating a sum of absolute differences of consecutive scene frames.

4. The method of claim 3, wherein the evaluating includes performing a scene change analysis.

5. The method of claim 4, wherein the evaluating, establishing, and adjusting produce accessible historical data.

6. The method of claim 5, wherein the historical data is referenced to perform subsequent renderings.

18. The method of claim 1, wherein the establishing includes an analysis of frame complexity.

19. The method of claim 18, wherein the dynamically adjusting includes at least one from the group including (i) adjusting the boundaries based upon the frame complexity and (ii) changing GPU rendering ratios based upon the frame complexity.

20. The method of claim 1, wherein the establishing includes an analysis of rendering prediction data.

7. A computer readable media storing instructions wherein said instructions when executed are adapted to process work items on a processing unit to perform a method for rendering a scene across N number of processors, wherein the method comprises:

evaluating performance statistics for each of the processors;

establishing load rendering boundaries for each of the processors, the boundaries defining a respective portion of the scene; and

dynamically adjusting the boundaries based upon the establishing and the evaluating.

8. The computer readable media of claim 7, wherein the processors are graphics processing units (GPUs).

9. The computer readable media of claim 7, wherein the evaluating includes calculating a sum of absolute differences of consecutive scene frames.

10. The computer readable media of claim 9, wherein the evaluating includes performing a scene change analysis.

11. The computer readable media of claim 10, wherein the evaluating, establishing, and adjusting produce accessible historical data.

12. The computer readable media of claim 11, wherein the historical data is referenced to perform subsequent renderings.

13. A computing system, comprising:

a system memory;

two or more processors coupled to the system memory; and

an execution unit configured to

evaluate performance statistics for each of the processors;

establish load rendering boundaries for each of the processors, the boundaries defining a respective portion of the scene; and

dynamically adjust the boundaries based upon the establishing and the evaluating.

14. The computing system of claim 13, wherein the processors are graphics processing units (GPUs).

15. The computing system of claim 13, wherein the processors include at least one from the group including graphics processing units (GPUs) and central processing units (CPUs).

16. The computing system of claim 13, wherein the evaluating includes determining frame complexity.

17. The computing system of claim 16, wherein the dynamically adjusting includes at least one from the group including (i) adjusting the boundaries based upon the frame complexity and (ii) changing GPU rendering ratios based upon the frame complexity.