Self-tuning statistical resource allocation for multipoint network events

• US 7,054,933 B2
• Filed: 03/19/2001
• Issued: 05/30/2006
• Est. Priority Date: 03/19/2001
• Status: Expired due to Term

• Current Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Original Assignee: Polycom Incorporated (HP Inc.)
• Inventors: Nylander, Eric Jay, Baxley, Warren Edward, Adams, Jeffrey Charles, Harbert, Douglas Earl
• Primary Examiner(s): LIN, WEN TAI
• Application Number:
  US09/812,971
• Publication Number:
  US 20020165963A1
• Time in Patent Office: 1,898 Days
• Field of Search: 709/200, 709204-205, 709/223, 709226-229, 718/104, 718/102, 705/1, 705 7- 8
• US Class Current: 709/226
• CPC Class Codes: H04L 12/1818 Conference organisation arrangements, e.g. handling schedules, setting up parameters needed by nodes to attend a conference, booking network resources, notifying involved parties
• CPC Class Codes: H04L 12/1827 Network arrangements for conference optimisation or adaptation
• CPC Class Codes: H04L 41/142 using statistical or mathematical methods
• CPC Class Codes: H04L 41/147 for predicting network behaviour
• CPC Class Codes: H04L 47/15 in relation to multipoint traffic arrangements for broadcast or multicast in data networks H04L12/18
• CPC Class Codes: H04L 47/70 Admission control; Resource allocation
• CPC Class Codes: H04L 47/801 Real time traffic
• CPC Class Codes: H04L 47/806 Broadcast or multicast traffic
• CPC Class Codes: H04L 47/822 Collecting or measuring resource availability data
• CPC Class Codes: H04L 47/826 Involving periods of time
• CPC Class Codes: H04L 47/83 based on usage prediction
• CPC Class Codes: H04L 65/1101 Session protocols
• CPC Class Codes: H04L 65/4038 with floor control
• CPC Class Codes: H04L 9/40 Network security protocols
• CPC Class Codes: H04M 3/36 Statistical metering, e.g. recording occasions when traffic exceeds capacity of trunks
• CPC Class Codes: H04M 3/567 Multimedia conference systems
• CPC Class Codes: H04N 7/152 Multipoint control units therefor

Abstract

A method for dynamically allocating MCU resources during a multipoint network event such as a conference call. The method determines the number of MCU resources to allocate for the start of the multipoint network event and then at each of a plurality of modeling intervals during the multipoint event adjusts the number of allocated MCU resources based upon actual inbound users. Self-tuning of the allocation of MCU resources for multipoint events occurs in advance of use by providing a look ahead allocation of resources based on what is likely to be needed in the future for a conferencing event. The number of multipoint events occurring within a tuning interval are counted. The number of MCU resources actually utilized during each multipoint event are accumulated and then a probability value is determined for future use of MCU resources for an upcoming multipoint event.

Claims

1. A method for allocating MCU ports for a multipoint network event, said method comprising:

receiving an allocation request for the multipoint network event, said request comprises a maximum number of MCU ports for the multipoint network event; and

determining electronically the number of MCU parts to allocate at the start of the multipoint network event, by calculating;

R=[(R_MAX)(R_SP)]_RSM, where R=number of ports to start, R_MAX=maximum ports, R_SP=configurable start port percentage, and R_SM=configurable minimum start ports, wherein the start MCU ports allocation number is less than or equal in value to the maximum MCU ports number.

2. The method of claim 1 wherein the allocation request is transmitted through a plurality of MCUs.

3. The method of claim 1 wherein the allocation request is transmitted through a common channel signaling interface.

4. The method of claim 1 wherein the step of receiving the allocation request originates as an external allocation request.

5. A method for time varying allocation of MCU ports during a multipoint network event, said method comprising:

determining the number of MCU ports to allocate for the start of the multipoint network event, by calculating;

R=[(R_MAX)(R_SP)]_RSM, where R=number of ports to start, R_MAX=maximum ports, R_SP=configurable start port percentage, and R_SM=configurable minimum start ports; and

at each of a plurality of modeling intervals during the multipoint network event, dynamically adjusting the number of allocated MCU ports based on users actually in the multipoint network event and based on a statistics algorithm using probability values related to future or historical use of MCU ports, wherein the probability values are dynamically modified.

6. The method of claim 5, wherein the step of adjusting comprises the step of calculating:

21. The method of claim 6, wherein the value for P_j comprises calculating:

7. A method for allocating resources for a multipoint network event, said method comprising:

obtaining available MCU capacity in a plurality of MCUs;

receiving an allocation request from an allocation requestor for the multipoint network event;

determining the number of resources to allocate to start the multipoint network event, wherein said determination comprises calculating;

R=[(R_MAX)(R_SP)]_RSM, where R=said number of resources to start, R_MAX=maximum resources, R_SP=configurable start resource percentage, and R_SM=configurable minimum start resources;

allocating the number of resources to at least one MCU;

debiting the allocated resources from the available MCU capacity; and

directing inbound users to the at least one MCU for participation in the multipoint network event.

8. The method of claim 7 wherein the allocation request is transmitted through a plurality of MCUs.

9. The method of claim 7 wherein the allocation request is transmitted through a common channel signaling interface.

10. The method of claim 7 wherein the step of receiving the allocation request originates as an external allocation request.

11. The method of claim 7 further comprising:

at each of a plurality of modeling intervals, dynamically adjusting the number of resources based on the number of inbound users actually in the multipoint network event and based on a statistics algorithm using probability values related to future or historical use of MCU capacity, wherein the probability values are dynamically modified.

12. The method of claim 11 wherein the step of adjusting comprises the step of calculating:

13. The method of claim 12 wherein the step of adjusting adjusts the number of resources at each modeling interval (j) based on a configurable value for S.

14. The method of claim 12 wherein the step of adjusting adjusts the number of resources at each modeling interval (j) is based on a self-tuning value for P_j.

15. The method of claim 14 wherein the self-tuning value for P_j comprises the step of calculating:

16. A method for allocating MCU ports for a plurality of multipoint network events, said method comprising:

obtaining available MCU capacity in a plurality of MCUs;

receiving allocation requests from allocation requestors for the plurality of multipoint network events;

determining based on each of said plurality of received allocation requests, the number of ports to allocate to start each of the multipoint network events, by calculating;

R=[(R_MAX)(R_SP)]_RSM, where R=number of ports to start. R_MAX=maximum ports, R_SP=configurable start port percentage, and R_SM=configurable minimum start ports;

allocating the number of ports to at least one MCU of said plurality of MCUs for each of the multipoint network events; and

at each of a plurality of modeling intervals during each multipoint network event, dynamically adjusting the number of ports based on inbound users actually in the multipoint network event and based on a statistics algorithm using probability values related to future or historical use of MCU ports, wherein the probability values are dynamically modified.

17. A method for tuning the allocation of MCU resources for multipoint network events, said method comprising:

counting the number of multipoint network events that have been accumulated within at least one predetermined tuning interval,

normalizing MCU resources actually utilized during each multipoint network event in each of the at least one predetermined tuning interval,

determining a probability value for future use of MCU resources for an upcoming multipoint network event based on the steps of counting and normalizing.

18. The method of claim 17 wherein the step of normalizing accumulates the number of resources at predetermined modeling intervals during each multipoint event.

19. The method of claim 17 wherein the predetermined tuning interval is at least three orders of magnitude in time greater than the predetermined modeling interval.

20. The method of claim 19 wherein the tuning interval is at least a day and the modeling interval is at least a minute.