System and method for selectively consolidating applications to a machine using resource utilization data

• US 8,521,747 B2
• Filed: 07/06/2012
• Issued: 08/27/2013
• Est. Priority Date: 07/19/2011
• Status: Active Grant

• Current Assignee: Infosys Limited
• Original Assignee: Infosys Limited
• Inventors: Bose, Sumit Kumar, Malviya, Rashi
• Primary Examiner(s): Saeed, Usmaan
• Assistant Examiner(s): Weinrich, Brian E.
• Application Number:
  US13/543,251
• Publication Number:
  US 20130024455A1
• Time in Patent Office: 417 Days
• Field of Search: 718/105, 707/740
• US Class Current: 707/740
• CPC Class Codes: G06F 11/3442 for planning or managing the needed capacity
• CPC Class Codes: G06F 11/3466 Performance evaluation by tracing or monitoring
• CPC Class Codes: G06F 2201/865 Monitoring of software

Abstract

A system and method of selectively consolidating applications on a machine comprises retrieving resource utilization data for a plurality of applications over a plurality of cycles for designated resource dimensions. A representative utilization profile is calculated on each application for each resource dimension. Load mean and variance values of the applications are calculated for each resource dimension. Each application is assigned to a class based the mean and variance values. A consolidation algorithm is computed on each application for each resource dimension to determine if resource consumption of the consolidated applications potentially exceed an available capacity in a machine.

Claims

1. A method of selectively consolidating applications on a machine, the method comprising:

retrieving, from a database, resource utilization data for a plurality of applications running on one or more machines, wherein the resource utilization data gathered from the database is over a plurality of cycles in relation to one or more designated resource dimensions;

calculating, using a processor on a network device, a representative utilization profile for a first application for each of the one or more designated resource dimensions using a curve representation technique;

calculating, using the processor, a first load mean value and a first load variance value of the first application for each of the one or more resource dimensions and assigning the first application to a first class category based the first load mean value and the first load variance value;

selecting a second application assigned to the first class category to consolidate with the first application, wherein the second application has a first load mean value and a first load variance value based on a calculated representative utilization profile for the second application;

computing, using the processor, a consolidation algorithm on the first and second applications for a respective resource dimension of the one or more designated resource dimensions, wherein the processor, executing the consolidation algorithm, at least partially analyzes the first and second load mean values and the first and second variance values and to determine if resource consumption of the consolidated first and second applications would potentially exceed an available capacity in the one or more designated resource dimensions of a selected machine;

generating a candidate list record of one or more proposed applications to consolidate with the first application, wherein the second application is in the candidate list record;

selecting a third application in the same category with the first and second applications;

combining load mean values of the first and third applications;

combining load variance values of the first and third applications; and

adding the third application to the candidate list record if the combination of the load mean values is less than a capacity of the machine and the combination of load variance values of the first and third applications is less than the load variance value of the first application and the combination of load variance values of the first and third applications is less than the load variance value of the third application.

2. The method of claim 1, wherein the consolidation algorithm is a first fit algorithm.

3. The method of claim 1, wherein the consolidation algorithm is a best fit algorithm.

4. The method of claim 1, wherein the consolidation algorithm is a string matching algorithm.

5. The method of claim 1, further comprising:

iteratively computing the consolidation algorithm on the first and second applications for all remaining one or more designated resource dimensions, wherein the consolidation algorithm at least partially analyzes the first and second load mean values and the first and second variance values and to determine if resource consumption of the consolidated first and second applications would potentially exceed the available capacity in the remaining one or more designated resource dimensions of the selected machine; and

proposing consolidation of the first and second applications if the consolidation algorithm indicates that the resource consumption of the consolidated first and second applications would not exceed available capacities of the remaining one or more designated resource dimensions of the selected machine.

6. The method of claim 1 further comprising collecting the resource utilization data for the first, second and third applications over a user designated number of cycles.

7. A non-transitory processor readable medium having stored thereon instructions for selectively consolidating applications on a network device, comprising processor executable code which when executed by a processor of the network device, causes the processor to:

retrieve, from a database, resource utilization data for a plurality of applications running on one or more machines, wherein the resource utilization data gathered from the database is over a plurality of cycles in relation to one or more designated resource dimensions;

calculate a representative utilization profile for a first application for each of the one or more designated resource dimensions using a curve representation technique;

calculate a first load mean value and a first load variance value of the first application for each of the one or more resource dimensions and assign the first application to a first class category based the first load mean value and the first load variance value;

select a second application assigned to the first class category to consolidate with the first application, wherein the second application has a first load mean value and a first load variance value based on a calculated representative utilization profile for the second application;

compute a consolidation algorithm on the first and second applications for a respective resource dimension of the one or more designated resource dimensions, wherein the computer, executing the consolidation algorithm, at least partially analyzes the first and second load mean values and the first and second variance values and to determine if resource consumption of the consolidated first and second applications would potentially exceed an available capacity in the one or more designated resource dimensions of a selected machine;

generate a candidate list record of one or more proposed applications to consolidate with the first application, wherein the second application is in the candidate list record;

select a third application in the same category with the first and second applications;

combine load mean values of the first and third applications;

combine load variance values of the first and third applications; and

adding the third application to the candidate list record if the combination of the load mean values is less than a capacity of the machine and the combination of load variance values of the first and third applications is less than the load variance value of the first application and the combination of load variance values of the first and third applications is less than the load variance values of the third application.

8. The computer readable medium of claim 7, wherein the consolidation algorithm is a first fit algorithm.

9. The computer readable medium of claim 7, wherein the consolidation algorithm is a best fit algorithm.

10. The computer readable medium of claim 7, wherein the consolidation algorithm is a string matching algorithm.

11. The computer readable medium of claim 7, wherein the processor executable code which when executed by the processor, causes the processor to:

iteratively compute the consolidation algorithm on the first and second applications for all remaining one or more designated resource dimensions, wherein the consolidation algorithm at least partially analyzes the first and second load mean values and the first and second variance values and to determine if resource consumption of the consolidated first and second applications would potentially exceed the available capacity in the remaining one or more designated resource dimensions of the selected machine;

propose consolidation of the first and second applications if the consolidation algorithm indicates that the resource consumption of the consolidated first and second applications would not exceed available capacities of the remaining one or more designated resource dimensions of the selected machine.

12. The processor readable medium of claim 7, wherein the processor executable code which when executed by the processor, causes the processor to collect the resource utilization data for the first, second and third applications over a user designated number of cycles.

13. A network device comprising:

a network interface configured to allow communications with at least one database;

a memory;

a processor coupled to the network interface and the memory, the processor operative to;

retrieve, from the at least one database, resource utilization data for a plurality of applications running on one or more machines, wherein the resource utilization data gathered from the database is over a plurality of cycles in relation to one or more designated resource dimensions;

calculate a representative utilization profile for a first application for each of the one or more designated resource dimensions using a curve representation technique;

calculate a first load mean value and a first load variance value of the first application for each of the one or more resource dimensions and assign the first application to a first class category based the first load mean value and the first load variance value;

select a second application assigned to the first class category to consolidate with the first application, wherein the second application has a first load mean value and a first load variance value based on a calculated representative utilization profile for the second application;

compute a consolidation algorithm on the first and second applications for a respective resource dimension of the one or more designated resource dimensions, wherein the processor, executing the consolidation algorithm, at least partially analyzes the first and second load mean values and the first and second variance values and to determine if resource consumption of the consolidated first and second applications would potentially exceed an available capacity in the one or more designated resource dimensions of a selected machine;

generate a candidate list record of one or more proposed applications to consolidate with the first application, wherein the second application is in the candidate list record;

select a third application in the same category with the first and second applications;

combine load mean values of the first and third applications;

combine load variance values of the first and third applications; and

adding the third application to the candidate list record if the combination of the load mean values is less than a capacity of the machine and the combination of the load variance values of the first and third applications is less than the load variance value of the first application and the combination of load variance values of the first and third applications is less than the load variance value of the third application.

14. The network device of claim 13, wherein the consolidation algorithm is a first fit algorithm.

15. The network device of claim 13, wherein the consolidation algorithm is a best fit algorithm.

16. The network device of claim 13, wherein the consolidation algorithm is a string matching algorithm.

17. The network device of claim 13, wherein the processor is configured to:

iteratively compute the consolidation algorithm on the first and second applications for all remaining one or more designated resource dimensions, wherein the consolidation algorithm at least partially analyzes the first and second load mean values and the first and second variance values and to determine if resource consumption of the consolidated first and second applications would potentially exceed the available capacity in the remaining one or more designated resource dimensions of the selected machine; and

propose consolidation of the first and second applications if the consolidation algorithm indicates that the resource consumption of the consolidated first and second applications would not exceed available capacities of the remaining one or more designated resource dimensions of the selected machine.

18. The network device of claim 13, wherein the processor is configured to collect the resource utilization data for the first, second and third applications over a user designated number of cycles.