MAINTAINING APPLICATION OPERATIONS WITHIN A SUBOPTIMAL GRID ENVIRONMENT

• US 20090228892A1
• Filed: 05/21/2009
• Published: 09/10/2009
• Est. Priority Date: 01/14/2004
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: HAMILTON, RICK ALLEN II, JOSEPH, JOSHY, DI LUOFFO, VINCENT VALENTINO, FELLENSTEIN, CRAIG
• Granted Patent: US 8,136,118 B2
• Time in Patent Office: Days
• US Class Current: 718/104
• CPC Class Codes: G06F 9/5088 involving task migration

Abstract

A method, system, and program for maintaining application operations within a suboptimal grid environment are provided. An application profile expresses the operational requirements of an application across multiple heterogeneous resource platforms and expresses the priority of modular breakdown of an application so that usage of resources by the application can be adjusted when suboptimal conditions are detected for the application. The application is submitted to at least one resource node from among multiple resource nodes within a grid environment. Then, a service availability management agent monitors a performance status of the at least one resource node. The service availability management agent compares the performance status with an operational requirement specified for the platform of the at least one resource node in the application profile. If the performance status does not meet the operational requirement, then the service availability management agent adjusts the use by the application the resource nodes according to the application profile, such that the application continues to operate when suboptimal conditions arise in a grid environment.

Claims

1. A system for maintaining application operations within a suboptimal grid environment, comprising:

a grid environment comprising a plurality of computing systems each comprising at least one resource comprising at least one operating system, at least one processor, at least one file system, at least one database manager and at least one memory manager and communicatively connected over a network through a grid management system to share each said at least one resource through a plurality of web services comprising simple object access protocol, web services description language, and extensible mark-up language interfaces implemented within at least one web service layer extended by an open grid services architecture atop at least one grid service layer implemented within an open grid services architecture;

said grid management system for receiving a plurality of separate jobs from a plurality of client systems over said network;

said grid management system for accessing a profile stored as a document type definition of an extensible markup language expression for an application from among a plurality of applications triggered by a particular job from a grid application profiles database specifying a selection of at least one web service from among said plurality of web services and at least one grid service within said grid services layer required by said application;

said grid management system for querying a plurality of separate business grid management systems to determine which of said separate business grid management systems manages at least one resource node from among a plurality of resource nodes of said grid environment comprising said at least one resource, wherein said at least one resource node further comprises said at least one grid service required for said application in said profile and a price for each said at least one resource node;

said grid management system for managing distribution of each of said plurality of separate jobs to a separate selection of said at least one resource with said particular job submitted to at least one resource node from among a selection of said plurality of resource nodes returning availability to handle said particular job at a selected price;

said grid management system for submitting an application from among a plurality of applications to said at least one resource node, wherein each of said plurality of separate jobs requests at least one of said plurality of applications;

said grid management system for monitoring a performance status of said at least one resource node running said application according to said profile;

said grid management system for comparing said performance status with an operational requirement specified in a profile for said application for when said application is operating at said at least one resource node;

said grid management system, responsive to said performance status not meeting said operational requirement, for determining whether there is at least one other resource node from among said plurality of resource nodes within said grid environment that meets said operational requirement specified in said profile for said application for when said application is operating at said at least one other resource node;

said grid management system, responsive to determining there is said at least one other resource node that meets said operational requirement specified in said profile, for relocating said application to said at least one other resource node within said grid environment;

said grid management system, responsive to determining there is not said at least one other resource node that meets said operational requirements specified is said profile, for determining from said profile at least one module to first shutdown from among a plurality of modules of said application defined in said profile, wherein each of said plurality of modules is assigned a separate resource size requirement and a separate priority to be shut down in said profile;

said grid management system, responsive to determining from said profile said at least one module to first shutdown, for sending an extensible markup language message to said at least one resource node authorizing said at least one resource node to shutdown said at least one module, such that said application continues to operate with a portion of said plurality of modules when said performance status fails to meet said operational requirement in said profile.

said grid management system, responsive to determining there is not said at least one other resource node that meets said operation requirements specified in said profile, for determining whether said application can continue to operate without said at least one module;

said grid management system, responsive to a determination that said application cannot continue to operate without said first module, for returning an error message for said application to a particular client system requesting said application from among said plurality of client systems; and

said grid management system, responsive to a determination that said application can continue to operate with said first module, for determining whether a next module from among said plurality of modules should be identified to be shutdown if said performance status does not meet said operational requirement after shutting down said at least one module.

2. The system according to claim 1 for maintaining application operations within a suboptimal grid environment, wherein said at least one resource node operates on at least one platform for which said operational requirement is specified.

3. The system according to claim 1 for maintaining application operations wherein said application is one from among an stand-alone application, a service, an agent, and a controller operating within said grid environment.

4. A computer program product for maintaining application operations within a suboptimal grid environment residing on a computer readable medium comprising instructions which when executed on a computer system cause the computer system to:

enable a grid environment comprising a plurality of computing systems each comprising at least one resource comprising at least one operating system, at least one processor, at least one file system, at least one database manager and at least one memory manager and communicatively connected over a network through a grid management system to share each said at least one resource through a plurality of web services comprising simple object access protocol, web services description language, and extensible mark-up language interfaces implemented within at least one web service layer extended by an open grid services architecture atop at least one grid service layer implemented within an open grid services architecture;

receive a plurality of separate jobs from a plurality of client systems over said network at said grid management system;

access a profile stored as a document type definition of an extensible markup language expression for an application from among a plurality of applications triggered by a particular job from a grid application profiles database specifying a selection of at least one web service from among said plurality of web services and at least one grid service within said grid services layer required by said application;

query a plurality of separate business grid management systems to determine which of said separate business grid management systems manages at least one resource node from among a plurality of resource nodes of said grid environment comprising said at least one resource, wherein said at least one resource node further comprises said at least one grid service required for said application in said profile and a price for each said at least one resource node;

manage distribution of each of said plurality of separate jobs to a separate selection of said at least one resource with said particular job submitted to at least one resource node from among a selection of said plurality of resource nodes returning availability to handle said particular job at a selected price;

submit by said grid management system an application from among a plurality of applications to said at least one resource node, wherein each of said plurality of separate jobs requests at least one of said plurality of applications;

monitor by said grid management system a performance status of said at least one resource node running said application according to said profile;

compare by said grid management system said performance status with an operational requirement specified in a profile for said application for when said application is operating at said at least one resource node;

responsive to said performance status not meeting said operational requirement, determine by said grid management system whether there is at least one other resource node from among said plurality of resource nodes within said grid environment that meets said operational requirement specified in said profile for said application for when said application is operating at said at least one other resource node;

responsive to determining there is said at least one other resource node that meets said operational requirement specified in said profile, relocate by said grid management system said application to said at least one other resource node within said grid environment;

responsive to determining there is not said at least one other resource node that meets said operational requirements specified is said profile, determine by said grid management system from said profile at least one module to first shutdown from among a plurality of modules of said application defined in said profile, wherein each of said plurality of modules is assigned a separate resource size requirement and a separate priority to be shut down in said profile;

responsive to determining from said profile said at least one module to first shutdown, send an extensible markup language message by said grid management system to said at least one resource node authorizing said at least one resource node to shutdown said at least one module, such that said application continues to operate with a portion of said plurality of modules when said performance status fails to meet said operational requirement in said profile;

responsive to determining there is not said at least one other resource node that meets said operation requirements specified in said profile, determine by said grid management system whether said application can continue to operate without said at least one module;

responsive to a determination that said application cannot continue to operate without said first module, return by said grid management system an error message for said application to a particular client system requesting said application from among said plurality of client systems; and

responsive to a determination that said application can continue to operate with said first module, determine by said grid management system whether a next module from among said plurality of modules should be identified to be shutdown if said performance status does not meet said operational requirement after shutting down said at least one module.