ESTIMATING FUTURE GRID JOB COSTS BY CLASSIFYING GRID JOBS AND STORING RESULTS OF PROCESSING GRID JOB MICROCOSMS

• US 20090259511A1
• Filed: 06/24/2009
• Published: 10/15/2009
• Est. Priority Date: 01/12/2005
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: FELLENSTEIN, CRAIG WILLIAM, HAMILTON, RICK ALLEN II, JOSEPH, JOSHY, SEAMAN, JAMES W.
• Granted Patent: US 8,396,757 B2
• Time in Patent Office: Days
• US Class Current: 705/8
• CPC Class Codes: G06F 9/5072 Grid computing
• CPC Class Codes: G06Q 10/06375 Prediction of business process outcome or impact based on a proposed change
• CPC Class Codes: G06Q 30/0247 Calculate past, present or future revenues
• CPC Class Codes: G06Q 30/0283 Price estimation or determination
• CPC Class Codes: G06Q 30/08 Auctions

Abstract

A method, system, and program for estimating future grid job costs by classifying grid jobs and storing results of processing grid job microcosms are provided. In general, a client side agent estimates future grid job costs by comparing a current grid job of a particular classification with a history of stored costs for other grid jobs of that customer of that particular classification. In particular, the grid client agent for a client system enabled to submit grid jobs to a grid provider that facilitates a grid environment, calculates a ratio of an application based metric to a grid provider metric for processing a particular grid job. Then, the grid client agent creates a table with an entry comparing the application based metric to a cost per grid provider metric for the grid provider based on the calculated ratio. Next, the grid client agent stores the table with the entry. Then, responsive to detecting a next grid job, the grid client agent estimates a cost for the grid provider to process the next grid job based on a particular number of application based metric operations required for the next grid job, translated by the ratio into the grid provider metric and multiplied by the cost per grid provider metric.

Claims

1. A system for estimating a future cost of a grid job, said system comprising:

a client system communicatively connected to a network, wherein said client system is enabled to submit a particular grid job to a grid provider that facilitates a grid environment for processing said particular grid job;

said client system further comprising;

means for calculating a ratio of an application based metric to a grid provider metric for said grid provider processing said particular grid job by;

distributing a job microcosm of said particular grid job from said client system to said grid provider via a network for processing in said grid environment; and

responsive to receiving a result of said job microcosm and a charge for processing said job microcosm based on a grid provider metric from said grid provider, calculating at said client system said ratio of said application based metric to said grid provider metric and identifying said cost per grid provider metric from said charge for processing said job microcosm;

means for creating a table with an entry comparing said application based metric to a cost per grid provider metric for said grid provider with said ratio;

means for storing said table with said entry for said particular grid job;

means, responsive to said client system detecting a next grid job classified by said at least one application metric, for accessing said entry comprising said application based metric to said cost per grid provider metric for said grid provider with said radio; and

means for calculating an estimated cost for said grid provider to process a next grid job based on a particular number of application based metric operations required for said next grid job translated by said ratio into said grid provider metric and multiplied by said cost per grid provider metric.

2. The system according to claim 1 for estimating a future cost of a grid job, said client system further comprising:

means for distributing a job request specifying requirements for processing said job microcosm to said grid provider; and

means, responsive to receiving a bid for said grid provider to process said job microcosm according to requirements, for distributing said job microcosm to said grid provider for processing.

3. The system according to claim 1 for estimating a future cost of a grid job, wherein said means for creating a table with an entry comparing said application based metric to a cost per grid provider metric for said grid provider with said ratio, further comprises:

means for accessing said cost per grid provider metric from at least one from among charges received from said grid provider for processing said particular grid job and a published rate per grid provider metric.

4. The system according to claim 1 for estimating a future cost of a grid job, said client system further comprising:

means, responsive to detecting at said client system an adjusted cost per grid provider metric for said grid provider, for updating said cost per grid provider metric in said table; and

means for automatically reestimating said cost for said grid provider to process said next grid job based on said particular number of application based metric operations required for said next job translated by said ratio into said grid provider metric and multiplied by said adjusted cost per grid provider metric.

5. The system according to claim 1 for estimating a future cost of a grid job, said client system further comprising:

means for submitting a plurality of job microcosms of said particular grid job to a plurality of grid providers;

means, responsive to receiving a plurality of results and a plurality of charges for processing said plurality of grid microcosms from each of said plurality of grid providers, for calculating a plurality of ratios of said application based metric to each of said plurality of grid providers metrics for processing said particular grid job;

means for creating said table with a plurality of entries, where each entry compares said application based metric to a cost for one of said plurality of grid providers per each of said plurality of grid provider metrics a particular ration for said one of said plurality of grid providers;

means for storing each of said plurality of entries in said table;

means for estimating a plurality of costs based on said application based metric for each of said plurality of grid providers to process said next job; and

means for comparing said plurality of costs to identify a least expensive grid provider from among said plurality of grid providers.

6. The system according to claim 1 for estimating a future cost of a grid job, said clients system further comprising:

means for classifying said particular grid job into a particular category from among a plurality of categories of grid jobs, wherein said particular category comprises grid jobs with operations measurable by said application based metric;

means for storing said entry for said particular grid job according to said particular category;

means, responsive to detecting a next grid job, for classifying said next grid job according to said particular category of grid job;

means for searching said table for said entry associated with said same particular category; and

means for estimating a cost for said grid provider to process said next grid job in said same particular category as said particular grid job based on said entry calculated based on said particular grid job.

7. A program for estimating a future cost of a grid job, said program embodied in a volatile or non-volatile computer-readable medium, said program comprising computer-executable instructions which cause a computer to perform the steps of:

calculating a ratio of an application based metric to a grid provider metric for processing said particular grid job by a particular grid provider, wherein said grid provider facilitates a grid environment enabled to process said particular grid job by;

distributing a job microcosm of said particular grid job from said client system to said grid provider via a network for processing in said grid environment; and

responsive to receiving a result of said job microcosm and a charge for processing said job microcosm based on a grid provider metric from said grid provider, calculating at said client system said ratio of said application based metric to said grid provider metric and identifying said cost per grid provider metric from said charge for processing said job microcosm;

creating a table with an entry comparing said application based metric to a cost per grid provider metric for said grid provider with said ratio;

storing said table with said entry for said particular grid job;

responsive to detecting a next grid job classified by said at least one application metric, accessing said entry comprising said application based metric to said cost per grid provider metric for said grid provider with said radio; and

calculating an estimated cost for said grid provider to process said next grid job based on a particular number of application based metric operations required for said next grid job translated by said ratio into said grid provider metric and multiplied by said cost per grid provider metric.

8. The program according to claim 7 for estimating a future cost of a grid job, further comprising the steps of:

distributing a job request specifying requirements for processing said job microcosm to said grid provider; and

responsive to receiving a bid for said grid provider to process said job microcosm according to requirements, distributing said job microcosm to said grid provider for processing.

9. The program according to claim 7 for estimating a future cost of a grid job, wherein said step of creating a table with an entry comparing said application based metric to a cost per grid provider metric for said grid provider with said ratio, further comprises the step of:

accessing said cost per grid provider metric from at least one from among charges received from said grid provider for processing said particular grid job and a published rate per grid provider metric.

10. The program according to claim 7 for estimating a future cost of a grid job, further comprising the steps of:

responsive to detecting an adjusted cost per grid provider metric for said grid provider, updating said cost per grid provider metric in said table; and

automatically reestimating said cost for said grid provider to process said next grid job based on said particular number of application based metric operations required for said next job translated by said ratio into said grid provider metric and multiplied by said adjusted cost per grid provider metric.

11. The program according to claim 7 for estimating a future cost of a grid job, further comprising the steps of:

classifying said particular grid job into a particular category from among a plurality of categories of grid jobs, wherein said particular category comprises grid jobs with operations measurable by said application based metric;

storing said entry for said particular grid job according to said particular category;

responsive to detecting a next grid job, classifying said next grid job according to said particular category of grid job;

searching said table for said entry associated with said same particular category; and

estimating a cost for said grid provider to process said next grid job in said same particular category as said particular grid job based on said entry calculated based on said particular grid job.