APPLICATION DEPLOYMENT IN HETEROGENEOUS ENVIRONMENTS

• US 20140040880A1
• Filed: 08/02/2012
• Published: 02/06/2014
• Est. Priority Date: 08/02/2012
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Brownlow, Sean T., Faiman, Nathan G., Kalmbach, Michael T., Petri, John E., Wendzel, Kevin
• Granted Patent: US 8,972,956 B2
• Time in Patent Office: Days
• US Class Current: 717/177
• CPC Class Codes: G06F 11/3466 Performance evaluation by tracing or monitoring
• CPC Class Codes: G06F 11/3616 using software metrics
• CPC Class Codes: G06F 8/60 Software deployment

Abstract

Techniques are disclosed for evaluating an application to determine a processing environment suited for executing the application. A request may be received to execute the application. If execution data identified for the application do not satisfy a first set of criteria, the application is deployed for execution on a first processing environment. If the execution data satisfy the first set of criteria, the application is deployed for execution partially in time on a second processing environment, based on the execution data. The second processing environment provides a higher capability than the first processing environment in terms of at least one resource type.

Claims

1. A computer-implemented method to determine a processing environment suited for executing an application, the method comprising:

receiving a request to execute the application;

upon determining that execution data identified for the application do not satisfy a first set of criteria, deploying the application for execution on a first processing environment; and

upon determining that the execution data satisfy the first set of criteria, deploying the application for execution partially in time on a second processing environment, based on the execution data, wherein the second processing environment provides a higher capability than the first processing environment in terms of at least one resource type.

2. The computer-implemented method of claim 1, wherein deploying the application for execution partially in time on the second processing environment upon determining that the execution data satisfy the first set of criteria comprises:

upon determining that trial data identified for the application do not satisfy a second set of criteria, deploying the application for execution on one or more candidate processing environments selected from a plurality of available processing environments based on the execution data for the application; and

upon determining that trial data identified for the application satisfy the second set of criteria, deploying the application for execution partially in time on the second processing environment, the second processing environment being selected from the one or more candidate processing environments based on the trial data for the application.

3. The computer-implemented method of claim 2, wherein each of the first set of criteria and the second set of criteria includes at least one of a qualitative criterion and a quantitative criterion, wherein the application is partially executed on the second processing environment for a first time duration determined based on the execution data, and wherein the application is partially executed on the first processing environment for a second time duration different from the first time duration.

4. The computer-implemented method of claim 3, wherein the execution data characterizes resource demands of the application over time in terms at least one resource type and when executing in the first processing environment;

wherein the trial data characterizes performance of the application when executing in the one or more candidate processing environments;

wherein each available processing environment provides an enhanced capability relative to the first processing environment in terms of at least one resource type.

5. The computer-implemented method of claim 4, wherein the application is deployed based further on static data generated via analyzing code of the application and without executing the application, wherein the code comprises at least one of source code, object code, and machine code;

wherein the application is deployed based further on a current utilization of the second processing environment, wherein the current utilization characterizes a load of the second processing environment in terms of a resource type.

6. The computer-implemented method of claim 5, wherein the application is executed on the first processing environment to generate execution data for the application, and wherein the application is executed on the one or more candidate processing environments to generate trial data for the application, and wherein the method further comprises:

analyzing the execution data for the application to generate an execution plan for the application;

selecting the one or more candidate processing environments based on the execution plan generated for the application;

analyzing the trial data for the application to update the execution plan for the application; and

selecting the second processing environment based on the updated execution plan for the application.

7. The computer-implemented method of claim 6, wherein the execution plan is generated based further on an execution plan template from a requesting entity from which the request to execute the application is received;

wherein the one or more candidate processing environments are selected based further on a service level agreement from the requesting entity, wherein the second processing environment is selected based further on the service level agreement, wherein the service level agreement specifies what is acceptable to the requesting entities in terms of at least one of processing environments, resource types, and monetary costs;

wherein each resource type is selected from at least processor, memory, storage, network, graphics, and audio, wherein each execution of the application is responsive to a respect request to execute the application.

8. The computer-implemented method of claim 7, wherein the first processing environment includes a hardware component pertaining to a resource demand of the application, the resource demand being of a resource type, wherein the plurality of available processing environments include a first available processing environment and a second available processing environment, each having a respective component pertaining to the resource demand of the application and better fulfilling the resource demand than the component of the first processing environment, wherein the respective component of each of the first and second available processing environments is associated with a distinct vendor;

wherein the first processing environment is a general-purpose processing environment, wherein the second processing environment is a special-purpose processing environment, wherein each of the one or more candidate processing environments is a special-purpose processing environment, wherein each of the plurality of available processing environments is a special-purpose processing environment, wherein the first set of criteria is different from the second set of criteria, wherein each of the first set of criteria and the second set of criteria includes at least one of a qualitative criterion and a quantitative criterion;

wherein the method further comprises;

further updating the execution plan based on executing the application on the second processing environment; and

generating an output for the requesting entity, based on the further updated execution plan, wherein the generated output includes one or more suggestions to the requesting entity in terms of at least one of processing environments, resource types, and monetary costs, as pertaining to the application, wherein the generated output is sent to the requesting entity.

9. A computer program product to determine a processing environment suited for executing an application, the computer program product comprising:

a computer-readable storage medium having computer-readable program code embodied therewith, the computer-readable program code comprising;

computer-readable program code configured to receive a request to execute the application;

computer-readable program code configured to, upon determining that execution data identified for the application do not satisfy a first set of criteria, deploy the application for execution on a first processing environment; and

computer-readable program code configured to, upon determining that the execution data satisfy the first set of criteria, deploy the application by operation of one or more computer processors when executing the computer-readable program code and for execution partially in time on a second processing environment, based on the execution data, wherein the second processing environment provides a higher capability than the first processing environment in terms of at least one resource type.

10. The computer program product of claim 9, wherein the computer-readable program code configured to deploy the application for execution partially in time on the second processing environment upon determining that the execution data satisfy the first set of criteria comprises:

computer-readable program code configured to, upon determining that trial data identified for the application do not satisfy a second set of criteria, deploy the application for execution on one or more candidate processing environments selected from a plurality of available processing environments based on the execution data for the application; and

computer-readable program code configured to, upon determining that trial data identified for the application satisfy the second set of criteria, deploy the application for execution partially in time on the second processing environment, the second processing environment being selected from the one or more candidate processing environments based on the trial data for the application.

11. The computer program product of claim 10, wherein each of the first set of criteria and the second set of criteria includes at least one of a qualitative criterion and a quantitative criterion, wherein the application is partially executed on the second processing environment for a first time duration determined based on the execution data, and wherein the application is partially executed on the first processing environment for a second time duration different from the first time duration.

12. The computer program product of claim 11, wherein the execution data characterizes resource demands of the application over time in terms at least one resource type and when executing in the first processing environment;

wherein the trial data characterizes performance of the application when executing in the one or more candidate processing environments;

wherein each available processing environment provides an enhanced capability relative to the first processing environment in terms of at least one resource type.

13. The computer program product of claim 12, wherein the application is deployed based further on static data generated via analyzing code of the application and without executing the application, wherein the code comprises at least one of source code, object code, and machine code;

wherein the application is deployed based further on a current utilization of the second processing environment, wherein the current utilization characterizes a load of the second processing environment in terms of a resource type.

14. The computer program product of claim 13, wherein the application is executed on the first processing environment to generate execution data for the application, and wherein the application is executed on the one or more candidate processing environments to generate trial data for the application, and wherein the computer-readable program code further comprises:

computer-readable program code configured to analyze the execution data for the application to generate an execution plan for the application;

computer-readable program code configured to select the one or more candidate processing environments based on the execution plan generated for the application;

computer-readable program code configured to analyze the trial data for the application to update the execution plan for the application; and

computer-readable program code configured to select the second processing environment based on the updated execution plan for the application.

15. The computer program product of claim 14, wherein the execution plan is generated based further on an execution plan template from a requesting entity from which the request to execute the application is received;

wherein the one or more candidate processing environments are selected based further on a service level agreement from the requesting entity, wherein the second processing environment is selected based further on the service level agreement, wherein the service level agreement specifies what is acceptable to the requesting entities in terms of at least one of processing environments, resource types, and monetary costs;

wherein each resource type is selected from at least processor, memory, storage, network, graphics, and audio, wherein each execution of the application is responsive to a respect request to execute the application.

16. The computer program product of claim 15, wherein the first processing environment includes a hardware component pertaining to a resource demand of the application, the resource demand being of a resource type, wherein the plurality of available processing environments include a first available processing environment and a second available processing environment, each having a respective component pertaining to the resource demand of the application and better fulfilling the resource demand than the component of the first processing environment, wherein the respective component of each of the first and second available processing environments is associated with a distinct vendor;

wherein the first processing environment is a general-purpose processing environment, wherein the second processing environment is a special-purpose processing environment, wherein each of the one or more candidate processing environments is a special-purpose processing environment, wherein each of the plurality of available processing environments is a special-purpose processing environment, wherein the first set of criteria is different from the second set of criteria, wherein each of the first set of criteria and the second set of criteria includes a qualitative criterion and a quantitative criterion;

wherein the computer-readable program code configured to further comprises;

computer-readable program code configured to further update the execution plan based on executing the application on the second processing environment; and

computer-readable program code configured to generate an output for the requesting entity, based on the further updated execution plan, wherein the generated output includes one or more suggestions to the requesting entity in terms of at least one of processing environments, resource types, and monetary costs, as pertaining to the application, wherein the generated output is sent to the requesting entity.

17. A system to determine a processing environment suited for executing an application, the system comprising:

one or more computer processors;

a memory containing a program which, when executed by the one or more computer processors, is configured to perform an operation comprising;

receiving a request to execute the application;

upon determining that execution data identified for the application do not satisfy a first set of criteria, deploying the application for execution on a first processing environment; and

upon determining that the execution data satisfy the first set of criteria, deploying the application for execution partially in time on a second processing environment, based on the execution data, wherein the second processing environment provides a higher capability than the first processing environment in terms of at least one resource type.

18. The system of claim 17, wherein deploying the application for execution partially in time on the second processing environment upon determining that the execution data satisfy the first set of criteria comprises:

upon determining that trial data identified for the application do not satisfy a second set of criteria, deploying the application for execution on one or more candidate processing environments selected from a plurality of available processing environments based on the execution data for the application; and

upon determining that trial data identified for the application satisfy the second set of criteria, deploying the application for execution partially in time on the second processing environment, the second processing environment being selected from the one or more candidate processing environments based on the trial data for the application.

19. The system of claim 18, wherein the application is partially executed on the second processing environment for a first time duration determined based on the execution data, and wherein the application is partially executed on the first processing environment for a second time duration different from the first time duration.

20. The system of claim 19, wherein each of the first set of criteria and the second set of criteria includes at least one of a qualitative criterion and a quantitative criterion, wherein the execution data characterizes resource demands of the application over time in terms at least one resource type and when executing in the first processing environment;

wherein the trial data characterizes performance of the application when executing in the one or more candidate processing environments;

wherein each available processing environment provides an enhanced capability relative to the first processing environment in terms of at least one resource type.

21. The system of claim 20, wherein the application is deployed based further on static data generated via analyzing code of the application and without executing the application, wherein the code comprises at least one of source code, object code, and machine code;

wherein the application is deployed based further on a current utilization of the second processing environment, wherein the current utilization characterizes a load of the second processing environment in terms of a resource type.

22. The system of claim 21, wherein the application is executed on the first processing environment to generate execution data for the application, and wherein the application is executed on the one or more candidate processing environments to generate trial data for the application, and wherein the operation further comprises:

analyzing the execution data for the application to generate an execution plan for the application;

selecting the one or more candidate processing environments based on the execution plan generated for the application;

analyzing the trial data for the application to update the execution plan for the application; and

selecting the second processing environment based on the updated execution plan for the application.

23. The system of claim 22, wherein the execution plan is generated based further on an execution plan template from a requesting entity from which the request to execute the application is received;

wherein the one or more candidate processing environments are selected based further on a service level agreement from the requesting entity, wherein the second processing environment is selected based further on the service level agreement, wherein the service level agreement specifies what is acceptable to the requesting entities in terms of at least one of processing environments, resource types, and monetary costs;

wherein each resource type is selected from at least processor, memory, storage, network, graphics, and audio, wherein each execution of the application is responsive to a respect request to execute the application.

24. The system of claim 23, wherein the first processing environment includes a hardware component pertaining to a resource demand of the application, the resource demand being of a resource type, wherein the plurality of available processing environments include a first available processing environment and a second available processing environment, each having a respective component pertaining to the resource demand of the application and better fulfilling the resource demand than the component of the first processing environment, wherein the respective component of each of the first and second available processing environments is associated with a distinct vendor;

wherein the first processing environment is a general-purpose processing environment, wherein the second processing environment is a special-purpose processing environment, wherein each of the one or more candidate processing environments is a special-purpose processing environment, wherein each of the plurality of available processing environments is a special-purpose processing environment, wherein the first set of criteria is different from the second set of criteria, wherein each of the first set of criteria and the second set of criteria includes a qualitative criterion and a quantitative criterion;

wherein the operation further comprises;

further updating the execution plan based on executing the application on the second processing environment; and

generating an output for the requesting entity, based on the further updated execution plan, wherein the generated output includes one or more suggestions to the requesting entity in terms of at least one of processing environments, resource types, and monetary costs, as pertaining to the application, wherein the generated output is sent to the requesting entity.