Method and apparatus for managing workload across multiple resources

• US 7,669,087 B1
• Filed: 07/31/2006
• Issued: 02/23/2010
• Est. Priority Date: 07/31/2006
• Status: Active Grant

• Current Assignee: Oracle America, Inc. (Oracle Corporation)
• Original Assignee: Sun Microsystems Incorporated (Oracle Corporation)
• Inventors: Delfatti, Daniel, Opheim, Hans Ludvig, Johnsen, Bjørn Dag
• Primary Examiner(s): Beausoliel; Robert
• Assistant Examiner(s): Lottich; Joshua P
• Application Number:
  US11/497,022
• Time in Patent Office: 1,303 Days
• Field of Search: 714/7, 714/13, 714/47, 709/224
• US Class Current: 714/6.32
• CPC Class Codes: G06F 11/008 Reliability or availability analysis

Abstract

A method for operating a system comprising multiple resources. The method comprises identifying for each resource a set of one or more failure risks for that resource. For each identified failure risk, a likelihood of failure is combined with an expected resolution time to provide a risk weight for the identified failure risk. For each resource, the risk weights for each failure risk are accumulated to provide an accumulated risk weight for the resource. A resource manager provisions workload across the multiple resources based on the accumulated risk weights for each resource.

Claims

1. A method for operating a system comprising multiple resources, said method comprising:

identifying for each resource a set of one or more failure risks for that resource, using a computer system;

combining for each identified failure risk a likelihood of failure and an expected resolution time to provide a risk weight for the identified failure risk;

accumulating for each resource the risk weights for each failure risk for that resource to provide an accumulated risk weight for the resource; and

provisioning workload across the multiple resources based on the accumulated risk weights for each resource, by allocating higher priority tasks to resources with lower accumulated risk weights.

2. The method of claim 1, wherein said likelihood of failure is based on an annualised failure rate (AFR) value.

3. The method of claim 2, wherein said AFR value is supplemented by a predictive analysis based on observed errors or other symptom patterns in the system.

4. The method of claim 1, wherein said expected resolution time comprises the time to perform a reset or restart of the resource.

5. The method of claim 4, wherein said expected resolution time is used to on-line a cold spare.

6. The method of claim 1, wherein said expected resolution time comprises the time to obtain and replace a FRU.

7. The method of claim 1, further comprising updating the accumulated risk weight in response to a change in the expected resolution time for at least one risk failure.

8. The method of claim 1, wherein said identifying, combining and accumulating are performed by a box manager, and said method further comprises transmitting the accumulated risk weights from the box manager to a resource manager, wherein said provisioning is performed by the resource manager.

9. The method of claim 1, wherein the multiple resources are grouped into fault zones such that all resources in a fault zone are subject to a single point of failure, and wherein the method further comprises determining an accumulated risk weight for each fault zone as well as for the individual resources.

10. An apparatus comprising multiple resources, said apparatus comprising:

a processor;

a set of one or more identified failure risks for each resource, wherein each identified failure risk has an associated likelihood of failure and expected resolution time;

logic configured to accumulate for each resource a set of risk weights to provide an accumulated risk weight for the resource, wherein the set of risk weights is determined based on the likelihood of failure and expected resolution time for each identified failure risk; and

a resource manager for provisioning workload across the multiple resources based on the accumulated risk weight for each resource, by allocating higher priority tasks to resources with lower accumulated risk weights.

11. The apparatus of claim 10, wherein said likelihood of failure is based on an annualised failure rate (AFR) value.

12. The apparatus of claim 11, wherein said AFR value is supplemented by a predictive analysis based on observed errors or other symptom patterns.

13. The apparatus of claim 10, wherein said expected resolution time comprises the time to perform a reset or restart of the resource.

14. The apparatus of claim 13, wherein said expected resolution time is used to on-line a cold spare.

15. The apparatus of claim 10, wherein said expected resolution time comprises the time to obtain and replace a field replaceable unit (FRU).

16. The apparatus of claim 10, wherein said logic is configured to update the accumulated risk weight in response to a change in the expected resolution time for at least one risk failure.

17. The apparatus of claim 10, further comprising a box manager corresponding to one or more resources, wherein said box manager includes the set of one or more identified failure risks for each of said one or more resources and the logic for accumulating a risk weight for each of said one or more resources.

18. The apparatus of claim 17, wherein said box manager is configured to transmit the accumulated risk weights for the one or more resources to the resource manager.

19. The apparatus of claim 10, wherein the multiple resources are grouped into fault zones such that all resources in a fault zone are subject to a single point of failure, and wherein said logic is configured to determine an accumulated risk weight for each fault zone as well as for the individual resources.

20. A computer readable storage medium storing a computer program comprising program instructions, said instructions when loaded onto a system comprising multiple resources causing the system to perform a method comprising:

identifying for each resource a set of one or more failure risks for that resource;

combining for each identified failure risk a likelihood of failure and an expected resolution time to provide a risk weight for the identified failure risk;

accumulating for each resource the risk weights for each failure risk for that resource to provide an accumulated risk weight for the resource; and

provisioning workload across the multiple resources based on the accumulated risk weights for each resource, by allocating higher priority tasks to resources with lower accumulated risk weights.