Method, system and computer program for discovering inventory information with dynamic selection of available providers

• US 7,885,947 B2
• Filed: 05/31/2007
• Issued: 02/08/2011
• Est. Priority Date: 05/31/2006
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Michele, Pescuma, Scotti, Alessandro, Pichetti, Luigi
• Primary Examiner(s): Zeender; F. Ryan
• Assistant Examiner(s): SHAAWAT, MUSSA A
• Application Number:
  US11/755,966
• Publication Number:
  US 20080235225A1
• Time in Patent Office: 1,349 Days
• Field of Search: None
• US Class Current: 707/705
• CPC Class Codes: G06F 21/10 Protecting distributed programs or content, e.g. vending or licensing of copyrighted material protection in video systems or pay television H04N7/16 ; Digital rights management [DRM]
• CPC Class Codes: G06F 21/16 Program or content traceability, e.g. by watermarking
• CPC Class Codes: G06Q 10/087 Inventory or stock management, e.g. order filling, procurement or balancing against orders

Abstract

A solution (200) for discovering inventory information in a data processing system is proposed. For this purpose, a corresponding discovery request is submitted (A1) to an inventory tool (200); the discovery request specifies a selected query pattern for the desired inventory information (for example, all the files included in a specific directory). Multiple providers (210)—such as of the interactive, cached or monitor type—are available for executing the required discovery operation. A predictive model (220) is associated with each provider; the models are used to estimate (A2,A3a,A3b-A4) the expected performance of the different providers for discovering the inventory information (according to the selected query pattern). In this way, it is possible to select (A5-A9) the provider, among the available ones, which is best suited for this purpose. The selected provider is then invoked (A10-A12) to discover the inventory information matching the selected query pattern. At the same time, the result of the discovery is used to refine (A13-A14) the models of the providers, so as to improve their accuracy.

Claims

1. A method for discovering inventory information in a data processing system, wherein a plurality of different providers for discovering the inventory information is available, the method including the steps of:

receiving, by the system, a request for discovering selected inventory information based on a selected query pattern,

by an inventory tool on the system, estimating an indication of performance of each provider for discovering the selected inventory information, the performance being estimated according to a predictive model of the provider depending on the selected query pattern,

selecting one of the providers according to the corresponding estimated performance,

discovering the selected inventory information by means of the selected provider, and

updating at least part of the models according to a result of the discovery; and

detecting an idle condition of the system according to a measured workload thereof,

simulating the receiving of a dummy request for discovering further selected inventory information based on a further selected query pattern in response to the idle condition, the dummy request being associated with a further selected provider,

further discovering the further selected inventory information by means of the further selected provider, and

updating at least part of the models according to a result of the further discovery.

2. The method according to claim 1, wherein the estimated performance is indicative of a time required to discover the selected inventory information.

3. The method according to claim 1, wherein each model includes at least one statistic parameter depending on the results of previous discoveries for each query pattern, the step of updating including:

updating the at least one statistic parameter for the selected query pattern according to the result of the discovery.

4. The method according to claim 3, wherein each discovery involves scanning entities available on the system, the at least one statistic parameter being based on an average number of the scanned entities and/or on an average time required to access each scanned entity.

5. The method according to claim 1, wherein each model includes at least one configuration parameter depending on a configuration of the system, the method further including the steps of:

detecting-any change in the configuration of the system, and

updating the at least one configuration parameter in response to the change.

6. The method according to claim 1, further including the step of:

logging an indication of frequency of previous requests for each query pattern, the step of selecting being further performed according to the frequency of the selected query pattern.

7. The method according to claim 1, further including the step of:

determining an indication of confidence of the estimated performance of each model, wherein the confidence is determined according to an error between the estimated performance and the result of the previous discoveries for the selected query pattern, the step of selecting being further performed according to the confidence of the estimated performance.

8. The method according to claim 1, wherein the step of selecting further includes:

comparing the estimated performance of one of the providers eligible to be selected with the estimated performance of a current one of the provider selected for a last discovery for the selected query pattern, and

switching to the eligible provider or maintaining the current provider according to the comparison.

9. A computer program product in a non-transistory, computer-usable storage medium the computer program when executed on a data processing system causing the system to perform a method for discovering inventory information in the system, wherein a plurality of different providers for discovering the inventory information is available, the method including the steps of:

receiving, by the system, a request for discovering selected inventory information based on a selected query pattern,

by the system, estimating an indication of performance of each provider for discovering the selected inventory information, the performance being estimated according to a predictive model of the provider depending on the selected query pattern,

selecting one of the providers according to the corresponding estimated performance,

discovering the selected inventory information by means of the selected provider, and

updating at least part of the models according to a result of the discovery; and

detecting an idle condition of the system according to a measured workload thereof,

simulating the receiving of a dummy request for discovering further selected inventory information based on a further selected query pattern in response to the idle condition, the dummy request being associated with a further selected provider,

further discovering the further selected inventory information by means of the further selected provider, and

updating at least part of the models according to a result of the further discovery.

10. A discovery tool for discovering inventory information in a data processing system, wherein a plurality of different providers for discovering the inventory information is available, the discovery system including:

a memory for storing computer instructions;

a processor for executing the stored computer instructions;

a collector engine for receiving a request for discovering selected inventory information based on a selected query pattern,

a provider manager for estimating an indication of performance of each provider for discovering the selected inventory information, the performance being estimated according to a predictive model of the provider depending on the selected query pattern, and

a selector for selecting one of the providers according to the corresponding estimated performance and for causing the selected provider to discover the selected inventory information, the provider manager updating at least part of the models according to a result of the discovery; and

wherein responsive to detecting an idle condition of the system according to a measured workload thereof, simulating the receiving of a dummy request for discovering further selected inventory information based on a further selected query pattern in response to the idle condition, the dummy request being associated with a further selected provider, causing the further selected provider to discover the further selected inventory information, and updating at least part of the models according to a result of the further discovery.