DISTRIBUTED DATA ANALYSIS

• US 20110055158A1
• Filed: 08/26/2009
• Published: 03/03/2011
• Est. Priority Date: 08/26/2009
• Status: Active Grant

• Current Assignee: Hewlett Packard Enterprise Development LP (Hewlett-Packard Enterprise Company)
• Original Assignee: Hewlett Packard Enterprise Development LP (Hewlett-Packard Enterprise Company)
• Inventors: Birdsall, David W., Gilbert, Gary M., Tapper, Gunnar D.
• Granted Patent: US 9,576,268 B2
• Time in Patent Office: Days
• US Class Current: 707/627
• CPC Class Codes: G06Q 10/10 Office automation; Time management
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising

Abstract

Distributed data analysis systems and methods are provided. A data broker distributes received data to consumers, such as information and repository consumers, which can be subscribed to the data. A subsystem with a processor for data processing can provide data to the data broker. A first information consumer may include a receiving module for receiving the data from the data broker, an analysis module for analyzing the received data to obtain a result, and a publication module for sending the result to the data broker when a second information consumer is subscribed to the result.

Claims

1. A computing system for distributed data analysis, comprising:

a data broker configured to distribute received data to information consumers subscribed to the received data;

a subsystem in communication with the data broker, comprising a processor for processing data and configured to provide raw data to the data broker; and

a first information consumer comprising;

a receiving module configured to receive the raw data from the data broker;

an analysis module configured to analyze the received raw data to obtain a analyzed result; and

a publication module configured to send the result to the data broker when a second information consumer is subscribed to the result.

2. A system in accordance with claim 1, further comprising:

a repository consumer configured to receive data from the data broker and to store the received data in a data repository; and

a computer readable medium in communication with the repository consumer and on which the data repository is stored.

3. A system in accordance with claim 1, wherein the analyzed result is represented in a same format as the raw data.

4. A system in accordance with claim 1, wherein the raw data comprises a first format and the system is configurable by using configuration data represented in the first format and through a subscription to the configuration data by at least one information consumer.

5. A system in accordance with claim 1, wherein the first information consumer is configured to pull data from the data repository and use the analysis module to analyze pulled data repository data together with the first data.

6. A system in accordance with claim 1, wherein the data broker comprises a notification module configured to notify a first information consumer to not send the second data to the data broker if a second information consumer is not subscribed to the second data.

7. A system in accordance with claim 1, further comprising a user interface consumer configured to provide system information to a user display device.

8. A system in accordance with claim 1, wherein loss of data sent to the data broker is signaled to the first information consumer and the first information consumer is configured to ignore the data loss when the lost data is unimportant to the analysis and to directly query the subsystem for the lost data when the lost data is important to the analysis.

9. A system in accordance with claim 1, wherein the first information consumer is configured to dynamically subscribe and unsubscribe to data from the data broker.

10. A system in accordance with claim 1, wherein the system is modular in that subsystems, data brokers, and consumers are addable and removable without reconfiguration of existing components of the system.

11. A system in accordance with claim 1, wherein the data broker comprises a cloud of data brokers.

12. A system in accordance with claim 1, wherein system components communicate via a messaging layer such that the components are platform independent.

13. A system in accordance with claim 1, wherein the subsystem further comprises a first subsystem configured to provide the first data to the data broker and a second subsystem configured to provide second data to the data broker, wherein the first data and the second data comprise different kinds of data and wherein the first information consumer is configured to subscribe to both the first data and the second data.

14. A system in accordance with claim 1, wherein the publication module is configured to send the analyzed result to the data broker as a plurality of kinds of data.

15. A method for distributed data analysis using a computing system, comprising:

sending data using a processor in a subsystem to a data broker;

distributing the data from the data broker to an information consumer subscribed to the data;

receiving the data at a first information consumer when the first information consumer is subscribed to the data, wherein the first information consumer analyzes the data to obtain an analyzed result and sends the analyzed result to the data broker when at least one of a second information consumer and a repository consumer is subscribed to the analyzed result; and

receiving at least one of the data and the analyzed result at the repository consumer from the data broker when the repository consumer is subscribed to the at least one of the data and the analyzed result, wherein the repository consumer stores the at least one of the data and the analyzed result in a data repository on a computer readable medium.

16. A method in accordance with claim 15, wherein the data comprises raw data, the method further comprising representing the analyzed result in a same format as the raw data.

17. A method in accordance with claim 15, wherein the data comprises a first format and the method further comprises configuring the system using configuration data represented in the first format and through a subscription to the configuration data by at least one information consumer.

18. A method in accordance with claim 15, wherein the data and the analyzed result comprise different kinds of data.

19. A method in accordance with claim 15, further comprising signaling loss of data to the first information consumer when data has been lost and determining whether to ignore the data loss or to query the subsystem to obtain the lost data.

20. A method for distributed data analysis using a computing system, comprising:

sending first data using a processor in a subsystem to a data broker;

distributing the first data from the data broker to a first information consumer subscribed to the first data;

receiving the first data at a first information consumer, wherein the first information consumer analyzes the first data to obtain a first result and sends the first result to the data broker in a same format as the first data, wherein the first result comprises less data than the first data;

distributing the first result from the data broker to a second information consumer subscribed to the first result, wherein the second information consumer analyzes the first result to obtain a second result and sends the second result to the data broker in a same format as the first data, wherein the second result comprises less data than the first result; and

storing at least one of the first data, the first result, and the second result in a data repository on a computer readable medium.