Capacity planning based on resource utilization as a function of workload

• US 8,402,468 B2
• Filed: 03/17/2008
• Issued: 03/19/2013
• Est. Priority Date: 03/17/2008
• Status: Active Grant

• Current Assignee: CA, Inc. (d/b/a CA Technologies) (Broadcom, Inc.)
• Original Assignee: CA, Inc. (d/b/a CA Technologies) (Broadcom, Inc.)
• Inventors: Seidman, David Isaiah, Addleman, Mark Jacob
• Primary Examiner(s): Puente, Emerson
• Assistant Examiner(s): SWIFT, CHARLES M
• Application Number:
  US12/049,840
• Publication Number:
  US 20090235268A1
• Time in Patent Office: 1,828 Days
• Field of Search: None
• US Class Current: 718/104
• CPC Class Codes: G06F 11/3442 for planning or managing the needed capacity
• CPC Class Codes: G06F 11/3452 Performance evaluation by statistical analysis
• CPC Class Codes: G06F 11/3466 Performance evaluation by tracing or monitoring
• CPC Class Codes: G06F 11/3476 Data logging G06F11/14, G06F11/2205 take precedence
• CPC Class Codes: G06F 2201/865 Monitoring of software

Abstract

Capacity planning based on resource utilization as a function of workload is disclosed. The workload may include different types of requests such as login requests, requests to visit web pages, requests to purchase an item on an online shopping site, etc. In one embodiment, data is determined for each of a plurality of workloads. The data includes characteristics of a workload and resource utilization due at least in part processing that workload. Based on the data, utilization of each of the resources as a function of workload characteristics is estimated. Further, based on the estimated resource utilization, workload characteristics that are expected to cause each respective resource to reach a certain level are predicted. That level could be 100 percent utilization, but could be another level. Capacity planning is performed based on the workload characteristics that are expected to cause each respective resource to reach a certain level.

Claims

1. A computer implemented method of predicting resource utilization, said method comprising:

determining a measure of requests of each of a plurality of request types in each of a plurality of workloads processed by a resource in a system;

determining a level of utilization of the resource for each workload, the utilization due at least in part to the resource processing the requests;

determining resource utilization as a function of the request types, the determining resource utilization as a function of the request types based on the level of utilization of the resource for each workload and the measure of requests of each of the request types in each workload; and

predicting utilization of the resource for a particular number of requests of each of the request types, the predicting based on the resource utilization as a function of the request types.

2. The method as recited in claim 1 wherein determining the resource utilization as a function of the request types comprises determining an equation that characterizes utilization of the resource as a function of the types of requests.

3. The method as recited in claim 2 further comprising applying the equation to determine a potential workload that is predicted to cause a particular level of utilization of the resource, the potential workload comprises a particular measure of at least one of the request types.

4. The method as recited in claim 3, wherein the particular level of utilization is 100 percent utilization.

5. The method as recited in claim 1, further comprising determining data that describes a level of utilization for at least one other resource due at least in part to the system processing each workload;

and further comprising, for each other resource;

determining, for each other resource, resource utilization as a function of the request types, the determining based on the data for each of the workloads; and

predicting utilization of each other resource for a particular measure of requests of each of the request types, the predicting based on the resource utilization as a function of the request types for each other resource.

6. The method as recited in claim 1 wherein the resource is a first resource and further comprising:

automatically determining which resources in the system are used to process at least one of the request types; and

selecting one of the resources that were determined to process the at least one of the request types as being the first resource.

7. The method as recited in claim 6 wherein the automatically determining which resources in the system are used to process at least one of the request types comprises inserting monitoring code into an application, said monitoring code provides information used to determine which resources in the system are used to process at least one of the request types.

8. The method as recited in claim 1 wherein the resource is a software resource.

9. The method of claim 1 wherein the determining resource utilization as a function of the request types comprises performing a regression analysis.

10. The method of claim 1 further comprising raising an alert if the predicted utilization of the resource for the particular measure of requests of each of the request types is at or above a certain level.

11. The method of claim 1 further comprising determining whether any of the resources are under-utilized based on the predicted utilization of the resource for a particular measure of requests of each of the request types.

12. A computer implemented method of capacity planning, the method comprising:

determining data for each of a plurality of workloads that are processed by a system having a plurality of resources, the data comprising;

a number of requests or rate of requests for each request type in each workload processed by the system; and

utilization of the plurality of resources due at least in part to the system processing each of the workloads;

performing the following for each of the plurality of resources;

based on the number or rate of requests for each of the request types in each workload, estimating utilization of a given resource as a function of the one or more request types; and

based on the estimated resource utilization as a function of the one or more request types, predicting a number or rate of requests for each of the request types that will cause a certain level of utilization of the given resource; and

performing capacity planning based on the number or rate of requests for each of the request types that will cause the certain level of utilization for each given resource.

13. A method as recited in claim 12 wherein the performing capacity planning comprises predicting which of the resources will be exhausted by the smallest workload.

14. A method as recited in claim 12 wherein the certain level of utilization is 100 percent utilization.

15. A method as recited in claim 12 wherein the certain level of utilization is less than 100 percent utilization.

16. The method as recited in claim 12 wherein the estimating utilization of a given resource is based on how many requests of each of the request types are in each workload processed by the system.

17. The method as recited in claim 12 wherein the estimating utilization of a given resource is based on a rate associated with requests of each of the request types in each workload.

18. The method as recited in claim 12 further comprising automatically identifying at least one of the plurality of resources that processes at least one of the request types prior to the determining data for each of a plurality of workloads.

19. The method as recited in claim 12 wherein the resources comprise software resources.

20. A non-transitory processor readable storage devices having processor readable code embodied thereon, wherein execution of the code on a processor causes the processor to:

determine a measure based on the number or rate of requests of each of a plurality of request types in each of a plurality of workloads processed by a resource in a system;

determine a level of utilization of the resource for each workload, the utilization due at least in part to the resource processing the requests;

determine a dependence of utilization of the resource upon each request type, the determining the dependence based on the level of utilization of the resource for each workload and the number or rate of requests of each of the request types in each workload; and

predict utilization of the resource for a particular number or rate of requests of each of the request types, the predicting based on the dependence of utilization of the resource upon each request type.

21. The non-transitory processor readable storage device of claim 20 wherein said determining a dependence of utilization comprises determining an equation that characterizes utilization of the resource as a function of the types of requests.

22. The non-transitory processor readable storage device of claim 21 wherein execution of the code on the processor further causes the processor to perform applying the equation to determine a potential workload that is predicted to cause a particular level of utilization of the resource, the potential workload comprises a particular number or rate of at least one of the request types.

23. The non-transitory processor readable storage device of claim 21 wherein the particular level of utilization is 100 percent utilization.

24. The non-transitory processor readable storage device of claim 20 wherein the processor further determines utilization of at least one other resource of the system due at least in part to the system processing each workload;

and wherein execution of the code on the processor further causes the processor to, for each other resource;

determine a dependence of utilization of each other resource upon each request type, the determining based on the system processing each workload; and

predict utilization of each other resource for a particular number or rate of requests of each of the request types, the predicting based on the dependence of utilization of each other resource upon each request type.

25. The non-transitory processor readable storage device of claim 20 wherein the resource is a first resource and wherein execution of the code on the processor further causes the processor to:

automatically determine which resources in the system are used to process at least one of the request types; and

select one of the resources that were determined to process the at least one of the request types as being the first resource.

26. The non-transitory processor readable storage device of claim 25 wherein the automatically determining comprises inserting monitoring code into an application, said monitoring code provides information used to determine which resources in the system are used to process at least one of the request types.

27. The non-transitory processor readable storage device of claim 20 wherein the resources comprise software resources.

28. The non-transitory processor readable storage device of claim 20 wherein the determining a dependence comprises performing a regression analysis.

29. The non-transitory processor readable storage device of claim 20 wherein execution of the code on the processor further causes the processor to perform raising an alert if the predicted utilization of the resource for the particular number of requests of each of the request types is at or above a certain level.

30. The non-transitory processor readable storage device of claim 20 wherein execution of the code on the processors further causes the processor to determine whether any of the resources are under-utilized based on the predicted utilization of the resource for a particular number of requests of each of the request types.

31. A non-transitory processor readable storage device having processor readable code embodied thereon, wherein execution of the code on a processor causes the processor to:

determine data for each of a plurality of workloads that are processed by a system having a plurality of resources, the data comprising;

a number of requests or rate of requests for each request type in each workload processed by the system; and

utilization of the plurality of resources due at least in part to the system processing each of the workloads;

perform the following for each of the plurality of resources;

based on the number or rate of requests for each of the request types in each workload, estimates utilization of a given resource as a function of the one or more request types; and

based on the estimated resource utilization as a function of the one or more request types, predict a number or rate of requests for each of the request types cause a certain level of utilization of the given resource; and

perform capacity planning based on the number or rate of requests for each of the request types that will cause the certain level of utilization for each given resource.

32. The non-transitory process or readable storage device of claim 31 wherein the performing capacity planning comprises predicting which of the resources will be exhausted by the smallest workload.

33. The non-transitory processor readable storage device of claim 31 wherein the certain level of utilization is 100 percent utilization.

34. A system for processing data, comprising:

computer readable storage having instructions stored therein; and

a processor coupled to the computer readable storage, said instructions cause said processor to implement;

a first module that determines how many requests of each of a plurality of request types are in each of a plurality of workloads processed by a resource in a system;

a second module that determines a level of utilization of the resource for each workload, the utilization due at least in part to the resource processing the requests;

a third module that determines resource utilization as a function of the request types, the determining the resource utilization as a function of the request types based on the level of utilization of the resource for each workload and how many requests of each of the request types are in each workload; and

a fourth module that predicts utilization of the resource for a particular number of requests of each of the request types, the predicting based on the resource utilization as a function of the request types.

35. The system of claim 34, wherein the third module determines the resource utilization as a function of the request types by determining an equation that characterizes utilization of the resource as a function of the types of requests.

36. The system of claim 35, wherein the fourth module applies the equation to determine a potential workload that is predicted to cause a particular level of utilization of the resource, the potential workload comprises a particular measure of at least one of the request types.