Granted Patent B1

US 8,677,003 · App. 11/974,113 · Granted Mar 18, 2014

Abstract

Claims (49)

1. A method for distributed processing of streaming call center data on an event protocol, comprising:

(a) receiving a plurality of related events from said streaming call center data at a node, said streaming call center data including data being dynamically collected during on-going phone calls while the data are being processed;

(b) amending a state of said related events based on a common key function of the related events;

(c) determining an error margin based on the amended state, said error margin representing an uncertainty in the streaming data as a result of a portion of said data being from on-going phone calls; and

(d) updating a current data transformation based on the amended state and error margin, thereby enabling real time analysis of streaming data.

2. The method of claim 1 , wherein said events are related by a key function.

3. The method of claim 2 , wherein said node is determined based on said key function.

4. The method of claim 1 , wherein said amending is based on a current state associated with said related events.

5. The method of claim 1 , wherein said updating includes modifying a prior measurement.

6. The method of claim 1 , wherein said updating includes modifying an interim result.

7. The method of claim 1 , wherein said updating includes:

(i) examining current states of a plurality of related events;

(ii) aggregating said current states; and

(iii) adjusting said aggregate with said error margin.

8. A server configured for distributed processing of streaming call center data on an event protocol, comprising:

a processor; and

a computer memory coupled to the processor, the computer memory including:

a data transformation module configured to receive a plurality of related events from said streaming call center data at a node and amend a state of said related events based on a common key function of the related events, said streaming call center data including data being dynamically collected during on-going phone calls while the data are being processed;

an error calculation module configured to determine an error margin based on the amended state, said error margin representing an uncertainty in the streaming data as a result of a portion of said data being from on-going phone calls; and

a result generation module configured to update a current data transformation based on the amended state and error margin, thereby enabling real time analysis of streaming data.

9. The server of claim 8 , wherein said events are related by a key function.

10. The server of claim 9 , wherein said node is determined based on said key function.

11. The server of claim 8 , wherein said amending is based on a current state associated with said related events.

12. A non-transitory computer-readable medium for distributed processing of streaming call center data on an event protocol, comprising logic instructions that, if executed:

(a) receive a plurality of related events from said streaming call center data at a node, said streaming call center data including data being dynamically collected during on-going phone calls while the data are being processed;

(b) amend a state of said related events based on a common key function of the related events;

(c) determine an error margin based on the amended state, said error margin representing an uncertainty in the streaming data as a result of a portion of said data being from on-going phone calls; and

(d) update a current data transformation based on the amended state and error margin, thereby enabling real time analysis of streaming data.

13. The computer-readable medium of claim 12 , wherein said events are related by a key function.

14. The computer-readable medium of claim 13 , wherein said node is determined based on said key function.

15. A method for distributed processing of streaming call center data on an event protocol, comprising:

(a) receiving a plurality of events from said streaming call center data at a first node, said streaming call center data including data being dynamically collected during on-going phone calls while the data are being processed;

(b) determining a second node for processing a set of related events; and

(c) sending said set of related events to said second node; said second node being configured to:

(1) amend a state of said related events based on a common key function of the related events;

(2) determine an error margin based on the amended state, said error margin representing an uncertainty in the streaming data as a result of a portion of said data being from on-going phone calls; and

(3) update a current data transformation based on the amended state and error margin, thereby enabling real time analysis of streaming data.

16. The method of claim 15 , wherein said events are related by a key function.

17. The method of claim 16 , wherein said second node is determined based on said key function.

18. The method of claim 15 , wherein said amending is based on a current state associated with said related events.

19. The method of claim 15 , wherein said updating includes modifying a prior measurement.

20. A system for distributed processing of streaming call center data on an event protocol, comprising:

a plurality of map nodes implemented in one or more computing devices for obtaining streaming call center data from a service, said streaming call center data including data being dynamically collected during on-going phone calls while the data are being processed;

a plurality of reduce nodes implemented in one or more computing devices and cross connected to said map nodes; each of said reduce nodes including:

(1) a data transformation module configured to receive a plurality of related events from said streaming data at a node and amend a state of said related events based on a common key function of the related events;

(2) an error calculation module configured to determine an error margin based on the amended state, said error margin representing an uncertainty in the streaming data as a result of a portion of said data being from on-going phone calls; and

(3) a result generation module configured to update a current data transformation based on the amended state and error margin, thereby enabling real time analysis of streaming data.

21. The system of claim 20 , wherein said events are related by a key function.

22. The system of claim 21 , wherein said second node is determined based on said key function.

Assignments (10)