METHOD AND APPARATUS FOR CAPACITY OPTIMIZATION AND PLANNING IN AN ON-DEMAND COMPUTING ENVIRONMENT

• US 20080183543A1
• Filed: 03/28/2008
• Published: 07/31/2008
• Est. Priority Date: 06/07/2005
• Status: Active Grant

• Current Assignee: Giuseppe Andrea Palcologo, Stephen Scott Briggs, Ta-Hsin Li
• Original Assignee: Giuseppe Andrea Palcologo, Stephen Scott Briggs, Ta-Hsin Li
• Inventors: Li, Ta-Hsin, Palcologo, Giuseppe Andrea, Briggs, Stephen Scott
• Granted Patent: US 8,286,175 B2
• Time in Patent Office: Days
• US Class Current: 705/8
• CPC Class Codes: G06F 2209/5012 Processor sets
• CPC Class Codes: G06F 9/4812 by interrupt, e.g. masked
• CPC Class Codes: G06Q 10/06312 Adjustment or analysis of established resource schedule, e.g. resource or task levelling, or dynamic rescheduling
• CPC Class Codes: G06Q 10/06315 Needs-based resource requirements planning or analysis
• CPC Class Codes: G06Q 10/06393 Score-carding, benchmarking or key performance indicator [KPI] analysis

Abstract

A method and apparatus are disclosed for determining the best cluster of computing resources to handle each of a plurality of workloads by optimally grouping multiple workloads and determining the best resource cluster to handle each group.

Claims

1. A method for optimizing computing resources used by on-demand customers, comprising the steps of:

storing in a computer database representations of a workload of each of a plurality of customers, each said workload being represented in terms of usage measures of a particular computing resource over a time period;

storing in a computer database, for each of said customer workloads, one or more parameters describing performance levels to be met by said particular computing resource, said performance levels being a function of said usage measures;

running an algorithm on said workloads, said algorithm optimizing usage of said particular computing resource, constrained by said performance levels, by grouping said plurality of customer workloads into groups, each group being served by a different cluster of one or more instances of said particular computing resource.

2. A method as in claim 1, wherein said particular computing resource is a central processing unit.

3. A method as in claim 1, wherein said parameters are obtained by displaying for a user a dialog specifying each of said parameters and receiving from said user a value for each of said specified parameters.

4. A method as in claim 3, wherein said specified parameters are upper and lower bounds and a capacity reserve level for performance of a service level agreement applicable to said customer with respect to said particular computing resource.

5. A method as in claim 1, wherein said algorithm is selected by a user from a dialog displaying alternatives for said optimizing.

6. A method as in claim 5, wherein said algorithm dialog displays a choice between hierarchical clustering and N-Box algorithms.

7. A method as in claim 6, wherein for each algorithm there is displayed a choice between optimizing the number of clusters or optimizing the total capacity of the instances of the particular computing resource.

8. An apparatus for optimizing computing resources used by on-demand customers, comprising:

means for storing in a computer database representations of a workload of each of a plurality of customers, each said workload being represented in terms of usage measures of a particular computing resource over a time period;

means for storing in a computer database, for each of said customer workloads, one or more parameters describing performance levels to be met by said particular computing resource, said performance levels being a function of said usage measures;

means for running an algorithm on said workloads, said algorithm optimizing usage of said particular computing resource, constrained by said performance levels, by grouping said plurality of customer workloads into groups, each group being served by a different cluster of one or more instances of said particular computing resource.

9. An apparatus as in claim 8, wherein said particular computing resource is a central processing unit.

10. An apparatus as in claim 8, wherein said parameters are obtained by displaying for a user a dialog specifying each of said parameters and receiving from said user a value for each of said specified parameters.

11. An apparatus as in claim 10, wherein said specified parameters are upper and lower bounds and a capacity reserve level for performance of a service level agreement applicable to said customer with respect to said particular computing resource.

12. An apparatus as in claim 8, wherein said algorithm is selected by a user from a dialog displaying alternatives for said optimizing.

13. An apparatus as in claim 12, wherein said algorithm dialog displays a choice between hierarchical clustering and N-Box algorithms.

14. An apparatus as in claim 13, wherein for each algorithm there is displayed a choice between optimizing the number of clusters or optimizing the total capacity of the instances of the particular computing resource.

15. A computer implemented system for optimizing computing resources used by on-demand customers, comprising:

first computer code for storing in a computer database representations of a workload of each of a plurality of customers, each said workload being represented in terms of usage measures of a particular computing resource over a time period;

second computer code for storing in a computer database, for each of said customer workloads, one or more parameters describing performance levels to be met by said particular computing resource, said performance levels being a function of said usage measures;

third computer code for running an algorithm on said workloads, said algorithm optimizing usage of said particular computing resource, constrained by said performance levels, by grouping said plurality of customer workloads into groups, each group being served by a different cluster of one or more instances of said particular computing resource.

16. A computer implemented system as in claim 15, wherein said particular computing resource is a central processing unit.

17. A computer implemented system as in claim 15, wherein said parameters are obtained by displaying for a user a dialog specifying each of said parameters and receiving from said user a value for each of said specified parameters.

18. A computer implemented system as in claim 15, wherein said algorithm is selected by a user from a dialog displaying alternatives for said optimizing.

19. A computer implemented system as in claim 18, wherein said algorithm dialog displays a choice between hierarchical clustering and N-Box algorithms.

20. A computer implemented system as in claim 19, wherein for each algorithm there is displayed a choice between optimizing the number of clusters or optimizing the total capacity of the instances of the particular computing resource.