Prediction and distribution of resource demand

• US 9,525,728 B2
• Filed: 09/17/2013
• Issued: 12/20/2016
• Est. Priority Date: 09/17/2013
• Status: Active Grant

• Current Assignee: Bank of America Corp.
• Original Assignee: Bank of America Corp.
• Inventors: Ardley, David, Sethi, Vince
• Primary Examiner(s): Shingles, Kristie
• Application Number:
  US14/029,383
• Publication Number:
  US 20150081877A1
• Time in Patent Office: 1,190 Days
• Field of Search: 709/226, 709/217, 709/219, 709/229
• US Class Current: 1/1
• CPC Class Codes: H04L 67/1012 based on compliance of requirements or conditions with available server resources
• CPC Class Codes: H04L 67/1031 Controlling of the operation of servers by a load balancer, e.g. adding or removing servers that serve requests

Abstract

A system for predicting one or more changes in demand for computational resources expected as a result of one or more events experienced by a business enterprise, and for selecting an appropriate distribution strategy for distributing computational tasks such that a single location with insufficient computational resources can communicate computational tasks in excess of a local computational resource capacity to one or more other locations for processing.

Claims

1. A system comprising:

a processor, configured to execute a first local distribution module;

a computer memory configured to store the first local distribution module; and

the first local distribution module in communication with a first server having a first level of computational resource availability, the first local distribution module further comprising;

a first event identification module, configured to;

search a received information stream for a keyword related to adverse weather conditions;

identify, based on the keyword being present within the received information stream, at least one geographic region predicted to experience an adverse weather event; and

a first predictive scaling engine, configured to;

identify a first amount of computational resources that will be requested from the first server based on the identified at least one geographic region predicted to experience an adverse weather event;

identify a second server in communication with the first local distribution module having a second level of computational resource availability; and

distribute to the second server, a first amount of computational tasks in excess of the first level of computational resource availability associated with the first server, for execution by the second server,

wherein the first local distribution module further comprises a local hardware lookup table storing the first level of computational resource availability associated with the first server and the second level of computational resource availability associated with the second server.

2. The system according to claim 1, further comprising:

a global distribution module in communication with the first local distribution module and a second local distribution module, the global distribution module having;

a global scaling engine, configured to;

identify a third server in communication with the second local distribution module; and

distribute, to the second local distribution module, a second amount of computational tasks in excess of the first level of computational resource availability associated with the first server, for execution by the third server,

wherein the processor is further configured to execute the global distribution module.

3. The system according to claim 2, wherein the global distribution module further comprises:

a data parsing module, configured to extract, from a plurality of data points stored in the first server, a sub-group of data points associated with the second amount of computational tasks, to reduce an amount of information to be communicated between the first and the third server.

4. The system according to claim 1, wherein the predictive scaling engine further comprises a reasoning engine configured to determine a response to the identified at least one geographic region predicted to experience an adverse weather event.

5. The system according to claim 2, wherein the global distribution module further includes a global hardware lookup table storing the first level of computational resource availability associated with the first server, the second level of computational resource availability associated with the second server, and a third level of computational resource availability associated with the third server.

6. A method executed by a processor performing steps comprising:

identifying, by a first event identification module, based on a keyword being present within an information stream received by a first local distribution module, at least one geographic region predicted to experience an adverse weather event, wherein the first local distribution module is in communication with a first server, wherein the first server has a first level of computational resource availability;

identifying, by a first predictive scaling engine, a first amount of computational resources that will be requested from the first server based on the identified at least one geographic region predicted to experience an adverse weather event;

identifying, by the first predictive scaling engine, a second server in communication with the first local distribution module, and having a second level of computational resource availability; and

distributing, by the first predictive scaling engine, a first amount of computational tasks in excess of the first level of computational resource availability associated with the first server, for execution by the second server,

wherein the first local distribution module further comprises a local hardware lookup table storing the first level of computational resource availability associated with the first server and the second level of computational resource availability associated with the second server.

7. The method of claim 6, wherein the first local distribution module and a second local distribution module are in communication with a global distribution module, and wherein the first and the second local distribution modules are in different geographic regions.

8. The method of claim 7, further comprising:

identifying, by a global scaling engine in communication with the global distribution module, a third server connected to the second local distribution module; and

distributing, by the global scaling engine, a second amount of computational tasks in excess of the first level of computational resource availability, for execution by the third server.

9. The method of claim 7, wherein the different geographic regions are in different time zones.

10. The method of claim 7, further comprising:

identifying the at least one geographic region predicted to experience an adverse weather event based on at least one predetermined rules stored in an event database.

11. The method of claim 7, wherein the identifying, by the first event identification module, further identifies an internal event associated with a business decision of a business enterprise.

12. The method of claim 7, wherein the identifying, by the first event identification module, further identifies an external event associated with at least one of a financial markets event, a political event, and a shipping event.

13. A non-transitory computer-readable storage medium having computer-executable program instructions store thereon that when executed by a processor cause the processor to perform steps comprising:

identifying, by a first event identification module of a first local distribution module, based on a keyword being present within a received information stream, at least one geographic region predicted to experience an adverse weather event from received data;

predicting a level of computational resources that will be requested from a first server in communication with the first local distribution module, based on the identified at least one geographic region predicted to experience an adverse weather event;

identifying a second server in communication with the first local distribution module; and

distributing a first amount of computational tasks in excess of a first level of computational resource availability associated with the first server, for execution by the second server,

wherein the first local distribution module further comprises a local hardware lookup table storing the first level of computational resource availability associated with the first server and a second level of computational resource availability associated with the second server.

14. The computer-readable storage medium of claim 13, wherein the computer-executable instructions, when executed by the processor, cause the processor to perform steps further comprising:

identifying a third server in communication with a second local distribution module; and

distributing to the second local distribution module, a second amount of computational tasks in excess of the first level of computational resource availability associated with the first server, for execution by the third server.

15. The computer-readable storage medium of claim 13, wherein the first local distribution module is in a first geographic location, and a second local distribution module is in a second geographic location.

16. The computer-readable storage medium of claim 15, wherein the first and second geographic locations are in different time zones.

17. The computer-readable storage medium of claim 13, wherein the identifying, by the first event identification module, further identifies an internal event based on a business decision of a business enterprise.

18. The computer-readable storage medium of claim 13, wherein the identifying, by the first event identification module, further identifies an external event based on at least one of a financial markets event, a political event, and a shipping event.

19. An apparatus comprising:

a global distribution module;

a first local distribution module, and a second local distribution module in communication with the global distribution module;

a first server in communication with the first local distribution module, and having a first level of computational resource availability;

a second server in communication with the second local distribution module, and having a second level of computational resource availability;

a first event identification module, configured to;

search a received information stream for a keyword related to adverse weather conditions;

identify, based on the keyword being present within the received information stream, at least one geographic region predicted to experience an adverse weather event; and

a first predictive scaling engine, configured to;

identify a first amount of computational resources that will be requested from the first server based on the identified at least one geographic region predicted to experience an adverse weather event; and

distribute to the second server, a first amount of computational tasks in excess of the first level of computational resource availability associated with the first server, for execution by the second server,

wherein the first local distribution module further comprises a local hardware lookup table storing the first level of computational resource availability associated with the first server and the second level of computational resource availability associated with the second server.