Device, system and method of managing a resource request

• US 20070157208A1
• Filed: 12/30/2005
• Published: 07/05/2007
• Est. Priority Date: 12/30/2005
• Status: Active Grant

• Current Assignee: Intel Corporation
• Original Assignee: Intel Corporation
• Inventors: Novakovsky, Larisa, Mandelblat, Julius, Mendelson, Abraham
• Granted Patent: US 7,958,510 B2
• Time in Patent Office: Days
• US Class Current: 718/104
• CPC Class Codes: G06F 2209/507 Low-level
• CPC Class Codes: G06F 9/5011 the resources being hardware resources other than CPUs, Servers and Terminals

Abstract

Embodiments of the present invention provide a resource management mechanism to monitor the availability of resources, detect the cause of a rejection, distinguish between different types of rejections, and manage the different types accordingly. For example, a queue manager in accordance with embodiments of the invention may be able to classify rejected requests, for example, as either a “long reject” or a “short reject” based on the cause of the rejection and the amount of time the rejection conditions are expected to remain valid. A short reject request may be rescheduled in an appropriate service queue, while a long reject request may be suspended in a reject queue. Other features are described and claimed.

Claims

1. A method comprising:

monitoring one or more rejection conditions for a request to access a resource; and

classifying a rejection of the request as either a long reject or a short reject based one or more parameters relating to the one or more rejection conditions.

2. The method of claim 1, wherein classifying a rejection of said request based on said one or more parameters comprises classifying based on a cause of the rejection and the amount of time said one or more rejection conditions are expected to remain valid.

3. The method of claim 1, wherein classifying a rejection comprises predicting the amount of time said one or more rejection conditions are expected to remain valid based on the cause of the rejection.

4. The method of claim 1, wherein monitoring said one or more rejection conditions comprises querying one or more resources to determine availability of said one or more resources.

5. The method of claim 1, wherein classifying a rejection comprises classifying a rejection as a short reject if said one or more rejection conditions include conditions dependent on an internal event.

6. The method of claim 1, wherein classifying a rejection comprises classifying a rejection as a long reject if said one or more rejection conditions include conditions dependent on an external resource.

7. The method of claim 1, comprising rescheduling said request in a service queue of said resource to be accessed if the request is rejected and classified as a short reject.

8. The method of claim 1, comprising suspending said request in a reject queue if the request is rejected and classified as a long reject.

9. The method of claim 8, comprising associating with said request in the reject queue data relating to the cause of the rejection and the resource to be accessed by said request.

10. The method of claim 9, comprising sending a release event to release said request from said reject queue based on said data and a signal from the resource to be accessed indicating that the resource is available to service the request.

11. The method of claim 8, comprising sending a release event to release said request from said reject queue based on one or more parameters indicative of a change in said one or more rejection conditions.

12. An apparatus comprising:

a resource management mechanism to monitor one or more rejection conditions for a request to access a resource and to classify a rejection of the request as either a long reject or a short reject based one or more parameters relating to the one or more rejection conditions.

13. The apparatus of claim 12, wherein said resource management mechanism is able to classify a rejection based on a cause of the rejection and the amount of time said one or more rejection conditions are expected to remain valid.

14. The apparatus of claim 12, wherein said resource management mechanism is able to predict the amount of time said one or more rejection conditions are expected to remain valid based on the cause of the rejection.

15. The apparatus of claim 12, wherein said resource management mechanism is able to query one or more resources to determine availability of said one or more resources.

16. The apparatus of claim 12, wherein said resource management mechanism is able to reschedule said request in a service queue of said resource to be accessed if the request is rejected and classified as a short reject.

17. The apparatus of claim 12, wherein said resource management mechanism is able to suspend said request if the request is rejected and classified as a long reject.

18. The apparatus of claim 17, wherein said resource management mechanism is able to associate with said request that is suspended data relating to the cause of the rejection and the resource to be accessed by said request.

19. The apparatus of claim 18, wherein said resource management mechanism is able to release said request that is suspended based on said data and a signal from the resource to be accessed indicating that the resource is available to service the request.

20. The apparatus of claim 17, wherein said resource management mechanism is able to release said request that is suspended based on one or more parameters indicative of a change in said one or more rejection conditions.

21. A system comprising:

a processor to monitor one or more rejection conditions for a request to access a resource and to classify a rejection of the request as either a long reject or a short reject based one or more parameters relating to the one or more rejection conditions; and

a memory to store one or more queues for management of the request according to whether the request is classified as a long reject or a short reject.

22. The system of claim 21, wherein said one or more queues include a service queue for the resource that is to be accessed by said request, and wherein said processor is able to reschedule the request in the service queue if the request is rejected and classified as a short reject.

23. The system of claim 21, wherein said one or more queues include a reject queue, and wherein said processor is able to suspend the request in the reject queue if the request is rejected and classified as a long reject.

24. The system of claim 23, wherein said processor is able to associate with said request in the reject queue data relating to the cause of the rejection and the resource to be accessed by the request, and wherein said memory is to store the data associated with the request.

25. The system of claim 24, wherein said processor is able to release said request from the reject queue based on said data and a signal from the resource to be accessed indicating that the resource is available to service the request.

26. The system of claim 24, wherein said processor is able to release said request from the reject queue based on one or more parameters indicative of a change in said one or more rejection conditions.

27. The system of claim 21, wherein said processor is able to query one or more resources to determine availability of said one or more resources.

28. The system of claim 21, wherein said processor is able to classify a rejection based on a cause of the rejection and the amount of time said one or more rejection conditions are expected to remain valid.

29. The system of claim 21, wherein said processor is able to predict the amount of time said one or more rejection conditions are expected to remain valid based on the cause of the rejection.