APPLICATION COMPATIBILITY IN MULTI-CORE SYSTEMS

• US 20090031318A1
• Filed: 07/24/2007
• Published: 01/29/2009
• Est. Priority Date: 07/24/2007
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Hsieh, Bor-ming, GOPALAN, YADHU, Miller, Mark
• Granted Patent: US 8,327,363 B2
• Time in Patent Office: Days
• US Class Current: 718/103
• CPC Class Codes: G06F 2209/483 Multiproc
• CPC Class Codes: G06F 2209/5018 Thread allocation
• CPC Class Codes: G06F 9/4881 Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues
• CPC Class Codes: G06F 9/5033 considering data affinity
• CPC Class Codes: G06F 9/52 Program synchronisation; Mutual exclusion, e.g. by means of semaphores

Abstract

Scheduling of threads in a multi-core system running various legacy applications along with multi-core compatible applications is configured such that threads from older single thread applications are assigned fixed affinity. Threads from multi-thread/single core applications are scheduled such that one thread at a time is made available to the cores based on the thread priority preventing conflicts and increasing resource efficiency. Threads from multi-core compatible applications are handled regularly.

Claims

1. A method to be executed at least in part in a computing device for managing application compatibility in a multi-core processor system, the method comprising:

receiving a thread from an application for processing by the processor system;

determining a type of the application; and

if the application is a multi-core compatible application, placing the thread in one of a global run queue and a per-processor queue based on a designation by the application and an affinity status of the thread.

2. The method of claim 1, further comprising:

if the application is a legacy application, placing a stub in one of the global run queue and a per-processor queue and scheduling threads from the legacy application one at a time through the stub.

3. The method of claim 2, further comprising:

modifying a position of the stub within the queue based on a priority of the legacy application thread and a priority of another thread in the queue.

4. The method of claim 3, wherein the legacy application thread has one of:

a high, a medium, and a low priority.

5. Tie method of claim 2, further comprising:

placing the stub in a per-processor queue if the legacy application has an affinity for one of a plurality of cores in the multi-core processor system.

6. The method of claim 1, wherein each core of the multi-core processing system is configured to check the global run queue and a per-processor queue associated with itself for a highest priority thread upon becoming available for processing.

7. The method of claim 6, wherein the legacy application and the multi-core compatible application are executed by one of:

the multi-core processing system locally and another processing system remotely.

8. The method of claim 1, further comprising:

providing a result of processing the thread to the submitting application.

9. The method of claim 1, wherein the thread is placed in one of the global run queue and the per-processor queue based on a priority assigned to the thread by the submitting application and a priority status of pending threads in the respective queues, and wherein the thread is picked based on a sequence number associated with a time of arrival of the thread if two threads in the global run queue and the per-processor queue have equal priority levels.

10. A system for managing application compatibility in a multi-core processor environment, the system comprising:

a memory;

a processor coupled to the memory, comprising a plurality of distinct processing cores, the processor configured to;

receive a thread from an application for processing;

determine a type of the application;

if the application is a multi-core compatible application;

place the thread in one of a global run queue and a per-processor queue based on a designation by the application and an affinity status of the thread; and

if the application is a legacy application;

place a stub in the global run queue; and

schedule the highest priority thread from the legacy application to be processed through the stub.

14. The system of claim 10, wherein the processor is configured to pick a thread based on a sequence number assigned by a scheduler if two threads in the global run queue and the per-processor queue have equal priorities.

15. The system of claim 10, wherein a distinct per-processor queue is reserved for each processing core.

16. The system of claim 10, wherein a priority of the thread for determining its place in the queue is based on at least one from a set of:

a priority designated by the submitting application, a time of arrival, a core availability, a number of pending threads, and a memory condition.

11. The system of claim 11, wherein the processor is further configured to schedule a next highest priority thread from the legacy application to be processed through the stub upon completion of processing of the previous thread.

12. The system of claim 11, wherein a position of the stub is adjusted based on a comparison of a priority level of a current thread from the legacy application and any other threads in the queue.

13. The system of claim 11, wherein the stub is placed in a per-processor queue if the legacy application designates a fixed affinity for its threads.

17. A computer-readable storage medium with instructions encoded thereon for managing application compatibility in a multi-core processor system, the instructions comprising:

receiving a thread from an application for processing by the processor system;

determining a type of the application;

if the application is a fixed affinity thread application;

placing the thread in a per-processor queue based on the assigned fixed affinity;

if the application is a legacy application;

determining an affinity status of the received thread;

placing a stub in one of the global run queue and a per-processor queue based on the affinity status;

scheduling threads from the legacy application one at a time through the stub; and

if the application is a multi-core compatible application;

placing the thread in one of a global run queue and a per-processor queue based on a designation by the application and an affinity status of the thread.

18. The computer-readable storage medium of claim 17, wherein the instructions further comprise:

assigning a sequence number to the thread based on a time of arrival of the thread;

picking the thread with the smaller sequence number for processing if two threads with equal priority are available in the global run queue and the per-processor queue for the particular core.

19. The computer-readable storage medium of claim 17, wherein a highest priority thread from the legacy application is placed in place of the stub upon completion of processing of a current thread from the same application.

20. The computer-readable storage medium of claim 19, wherein a position of the stub within the queue is adjusted based on a priority of a current thread in place of the stub and priorities of other threads in the queue.