Optimized usage of collector resources for performance data collection through even task assignment

• US 8,151,272 B2
• Filed: 04/07/2008
• Issued: 04/03/2012
• Est. Priority Date: 04/07/2008
• Status: Active Grant

• Current Assignee: AT&T Services Incorporated (AT&T, Inc.)
• Original Assignee: AT&T Intellectual Property I LP (AT&T, Inc.)
• Inventors: Bogdan, Pamela A., Gupta, Hema, Rosell, Mary Hahola, Sundaresan, Nagaraj
• Primary Examiner(s): An, Meng
• Assistant Examiner(s): Chew, Brian
• Application Number:
  US12/098,580
• Publication Number:
  US 20090254914A1
• Time in Patent Office: 1,457 Days
• Field of Search: 718/105
• US Class Current: 718/105
• CPC Class Codes: G06F 2209/5021 Priority
• CPC Class Codes: G06F 9/505 considering the load

Abstract

A method of balancing computer resources on a network of computers is provided employing a two-tier network architecture of at least one High Level Collector as a scheduler/load balancing server, and a plurality Low level Collectors which gather task data and execute instructions. Tasks are assigned priority and weight scores and sorted prior to assignment to Low Level Collectors. Also provided is a computer readable medium including instructions, wherein execution of the instructions by at least one computing device balances computer resources on a network of computers.

Claims

1. A method of balancing computer resources on a network of computers, comprising:

associating a plurality of tasks to a high level collector, priority score and weight score;

sorting the plurality of tasks into a list, the tasks being sorted first by priority score and then by weight score;

assigning the plurality of tasks into one of first and second queues, the first queue including highest weight scores sorted in descending order from highest to lowest weight, the second queue including lowest weight scores sorted in ascending order from lowest to highest weight, the quantity of tasks assigned to the first and second queues being substantially equal;

selecting tasks alternately from the first and second queues; and

assigning individual selected tasks to one of a plurality of low level collectors, the low level collectors individually having a total weight representing a sum of weight scores associated with tasks assigned thereto, the individual selected tasks being assigned to that low level collector not having reached capacity and having the lowest total weight just prior to assignment of respective ones of the individual selected tasks, the plurality of low level collectors comprising computer resources on a network of computers, the computer resources receiving tasks transmitted to the high level collector.

2. The method of claim 1, wherein the plurality of low level collectors have substantially non-equivalent processing power, wherein each low level collector has a pre-defined capacity limit comprising a maximum weight score, the method further comprising:

assigning tasks to the plurality of low level collectors until the capacity of one of the low level collectors is attained; and

assigning no additional tasks to any low level collector whose capacity limit would be exceeded if the next task would exceed the capacity limit thereon.

3. The method of claim 1, wherein a finite set of types of tasks is provided, and the weight and priority score for each type of task are dependent on a plurality of factors selected from a network protocol, a software request, and a hardware request.

4. The method of claim 1, wherein the weight score associated with any particular task is dependent on processing power and resource requirements.

5. The method of claim 1, wherein the priority score associated with any particular task is dependent on the urgency with which the task must be executed.

6. The method of claim 1, wherein the network of computers comprises a two-tiered architecture, wherein each high level collector assigns tasks to one or more low level collectors.

7. The method of claim 1, wherein the computer resources include servers or routers.

8. A method of balancing computer resources on a network of computers, comprising:

associating a plurality of tasks to a high level collector, priority score and a weight score;

sorting the plurality of tasks into a list, the tasks being sorted first by priority score and then by weight score;

assigning the plurality of tasks into one of first and second queues, the first queue including highest weight scores sorted in descending order from highest to lowest weight, the second queue including lowest weight scores sorted in ascending order from lowest to highest weight, the quantity of tasks assigned to the first and second queues being substantially equal;

selecting tasks alternately from the first and second queues; and

assigning individual selected tasks to one of a plurality of low level collectors, the low level collectors individually having a total weight representing a sum of weight scores associated with tasks assigned thereto, the individual selected tasks being assigned to that low level collector not having reached capacity and having the lowest total weight just prior to assignment of respective ones of the individual selected tasks, the plurality of low level collectors having substantially equivalent processing power, the plurality of tasks being assigned substantially evenly among the plurality of low-level collectors.

9. A non-transitory computer-readable medium storing instructions, wherein execution of the instructions by at least one hardware computing device balances computer resources on a network of computers by:

associating a plurality of tasks to a high level collector, priority score and weight score;

sorting the plurality of tasks into a list, the tasks being sorted first by priority score and then by weight score;

assigning the plurality of tasks into one of first and second queues, the first queue including highest weight scores sorted in descending order from highest to lowest weight, the second queue including lowest weight scores sorted in ascending order from lowest to highest weight, the quantity of tasks assigned to the first and second queues being substantially equal;

selecting tasks alternately from the first and second queues; and

assigning individual selected tasks to one of a plurality of low level collectors, the low level collectors individually having a total weight representing a sum of weight scores associated with tasks assigned thereto, the individual selected tasks being assigned to that low level collector not having reached capacity and having the lowest total weight just prior to assignment of respective ones of the individual selected tasks, the plurality of low level collectors comprising computer resources on a network of computers, the computer resources receiving tasks transmitted to the high level collector.

10. The non-transitory computer-readable medium storing instructions defined by claim 9, wherein the plurality of low-level collectors have substantially equivalent processing power, the plurality of tasks being assigned substantially evenly among the plurality of low-level collectors.

11. The non-transitory computer-readable medium storing instructions defined by claim 9, wherein the plurality of low level collectors have substantially non-equivalent processing power, wherein each low level collector has a pre-defined capacity limit comprising a maximum weight score, the method further comprising:

assigning tasks to the plurality of low level collectors until the capacity of one of the low level collectors is attained; and

assigning no additional tasks to any low level collector whose capacity limit would be exceeded if the next task would exceed the capacity limit thereon.

12. The non-transitory computer-readable medium storing instructions defined by claim 9, wherein a finite set of types of tasks is provided, and the weight and priority score for each type of task are dependent on a plurality of factors selected from a network protocol, a software request, and a hardware request.

13. The non-transitory computer-readable medium storing instructions defined by claim 9, wherein the weight score associated with any particular task is dependent on processing power and resource requirements.

14. The non-transitory computer-readable medium storing instructions defined by claim 9, wherein the priority score associated with any particular task is dependent on the urgency with which the task must be executed.

15. The non-transitory computer-readable medium storing instructions defined by claim 9, wherein the network of computers comprises a two-tiered architecture, wherein each high level collector assigns tasks to one or more low level collectors.

16. The non-transitory computer-readable medium defined by claim 9, wherein the computer resources include servers or routers.