METHOD AND SYSTEM FOR DECENTRALIZED WORKLOAD OPTIMIZATION IN A DATA PACKET PROCESSING SYSTEM USING A MULTICORE CPU

• US 20150200854A1
• Filed: 01/15/2015
• Published: 07/16/2015
• Est. Priority Date: 01/15/2014
• Status: Active Grant

• Current Assignee: Wind River Systems Incorporated (Aptiv PLC)
• Original Assignee: Wind River Systems Incorporated (Aptiv PLC)
• Inventors: BUCHNIK, Erez, MANDIN, Jeff
• Granted Patent: US 9,917,777 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 11/0724 in a multiprocessor or a multi-core unit multiprocessors per se G06F15/80
• CPC Class Codes: G06F 15/80 comprising an array of processing units with common control, e.g. single instruction multiple data processors G06F15/82 takes precedence ; for correlation function computation G06F17/15
• CPC Class Codes: G06F 9/5077 Logical partitioning of resources; Management or configuration of virtualized resources specific details on emulation or internal functioning of virtual machines G06F9/455
• CPC Class Codes: G06F 9/5083 Techniques for rebalancing the load in a distributed system
• CPC Class Codes: H04L 47/10 Flow control; Congestion control

Abstract

A system and method for dynamically processing data across a plurality of processor cores in a multi-core computing environment for breaking down incoming data traffic into a plurality of segments and dynamically distribute and re-distribute the segments on an ongoing basis so as to allocate the handling of data traffic in a resource efficient manner.

Claims

1. A method for dynamically rebalancing packet traffic workloads in a multi-core computing system, the system comprising a first processor core and a second processor core, the method comprising:

selecting a first traffic segment assigned to be processed by the first processor core, the first processor core having a first initial workload;

reassigning the first traffic segment so that it will be processed by the second processor core if an increased workload of the second processor core upon receipt of the first traffic segment will not exceed a first workload level; and

determining whether a reduced workload of the first processor core, subsequent to the transferring step, is above, at or below a second workload level.

2. The method of claim 1, further comprising selecting a second traffic segment configured to be processed by the first processor core and repeating the reassigning and determining steps for the second traffic segment if the reduced workload is above the second workload level.

3. The method of claim 1, wherein the first initial workload is above an upper threshold value corresponding to an excessively high processing load.

4. The method of claim 1, wherein the first and second workload levels both correspond to an optimal workload level.

5. The method of claim 1, wherein the first initial workload is below a lower threshold value corresponding to an excessively low processing load.

6. The method of claim 5, wherein the first workload level is an optimal workload level, and the second workload level is zero.

7. The method of claim 6, further comprising the step of inactivating the first processor core if the reduced workload is at zero.

8. A method for dynamically processing data across a plurality of processor cores in a computing system, the method comprising:

distributing a plurality of traffic segments across the plurality of processor cores, each of the plurality of traffic segments corresponding to a subset of incoming data received by the computing system;

executing the plurality of traffic segments via the plurality of processor cores;

determining whether an optimization counter exceeds a predetermined threshold;

re-distributing the plurality of traffic segments across the plurality of processor cores such that a workload of at least one of the processor cores is altered to approach a predetermined workload level, and resetting the optimization counter if the optimization counter exceeds the predetermined threshold.

9. The method of claim 8, wherein the predetermined workload level is an optimal workload level.

10. The method of claim 8, wherein the predetermined workload level is zero.

11. The method of claim 8, wherein the optimization counter corresponds to a number of computing cycles.

12. A computing system comprising:

a first processor core having a first initial workload, the first processor configured to execute a first and a second traffic segment on a first receiving queue, and to determine whether a first reduced workload of the first processor core upon removal of the first traffic segment, is above, at or below a first workload level; and

a second processor core configured to execute to the first traffic segment when a first increased workload of the second processor core upon receipt of the first traffic segment does not exceed a second workload level,

wherein when the first reduced workload is above the second workload level, the first processor core is further configured to determine whether a second reduced workload of the first processor core upon removal of the second traffic segment, is above, at or below the first workload level, and the second processor core is further configured to execute the second traffic segment when a second increased workload of the second processor core upon receipt of the second traffic segment does not exceed the second workload level.

13. The system of claim 12, wherein the first initial workload is above an upper threshold value corresponding to an excessively high processing load.

14. The system of claim 13, wherein the first and second workload levels both correspond to an optimal workload level.

15. The system of claim 12, wherein the first initial workload is below a lower threshold value corresponding to an excessively low processing load.

16. The system of claim 15, wherein the first workload level is an optimal workload level, and the second workload level is zero.

17. The system of claim 16, wherein the first processor core is further configured to be inactivated when either the first or second reduced workload is at zero.

18. A system comprising a plurality of processor cores and a set of instructions executing on the plurality of processor cores, the set of instructions being operable to:

distribute a plurality of traffic segments across the plurality of processor cores, each of the plurality of traffic segments corresponding to a subset of incoming data received by the computing system;

execute the plurality of traffic segments;

determine whether an optimization counter exceeds a predetermined threshold;

re-distribute the plurality of traffic segments across the plurality of processor cores such that a workload of at least one of the processor cores is altered to approach a predetermined workload level, and reset the optimization counter if the optimization counter exceeds the predetermined threshold.

19. The system of claim 18, wherein the predetermined workload level is an optimal workload level.

20. The system of claim 18, wherein the predetermined workload level is zero.

21. The system of claim 18, wherein the optimization counter corresponds to a number of computing cycles.