Dynamic allocation and configuration of a computer system's network resources

• US 20060029097A1
• Filed: 02/01/2005
• Published: 02/09/2006
• Est. Priority Date: 06/07/2004
• Status: Abandoned Application

• Current Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: McGee, Michael Sean, Enstone, Mark Richard, McIntyre, Michael Sean, Stratton, Mark Christopher, Howard, Gregory Thomas
• Application Number:
  US11/048,524
• Publication Number:
  US 20060029097A1
• Time in Patent Office: Days
• US Class Current: 370/468
• CPC Class Codes: H04L 47/10 Flow control; Congestion control
• CPC Class Codes: H04L 47/125 by balancing the load, e.g. traffic engineering
• CPC Class Codes: H04L 47/20 Traffic policing

Abstract

A resource allocation application is configured to run on a computer system that is coupled through a plurality of network resources to one or more networks. The resources are initially allocated among one or more teams and a pool. One or more usage policies are established for at least one of the teams. Resource usage is continuously monitored to identify actionable resource usage conditions. The network resources are automatically reconfigured in accordance with the one or more usage policies in response to the actionable resource usage conditions.

Claims

1. A method of dynamically allocating and configuring network resources of a computer system, said method comprising:

initially allocating the network resources among one or more teams and a pool;

establishing one or more usage policies for at least one of the one or more teams;

continuously monitoring usage of the network resources to identify actionable resource usage conditions; and

reconfiguring the network resources in accordance with the one or more usage policies in response to the actionable resource usage conditions.

2. The method of claim 1 further comprising preventing the computer system from using the pooled resources for purposes other than said pool.

3. The method of claim 1 wherein said reconfiguring the network resources further comprises adding network resources from the pool to at least one of the one or more teams in response to at least one of the identified actionable resource usage conditions.

4. The method of claim 1 wherein said reconfiguring the network resources further comprises recapturing network resources from at least one of the one or more teams to the pool in response to at least one of the identified actionable resource usage conditions.

5. The method of claim 4 wherein said recapturing network resources further comprises marking the recaptured resources as needing attention if they are not in a functional state.

6. The method of claim 5 wherein said recapturing network resources further comprises generating an alarm in response to resources needing attention being present in the pool.

7. The method of claim 6 wherein said recapturing network resources further comprises repairing or replacing resources needing attention and marking them available.

8. The method of claim 3 wherein the least one of the identified actionable resource usage conditions is that aggregated usage for the at least one team has reached a predetermined threshold for a predetermined period of time and the at least one team is a transmit load balancing (TLB) team.

9. The method of claim 1 wherein said reconfiguring the network resources further comprises changing the at least one team from a network fault tolerant (NFT) team to a TLB team when at least one of the identified actionable resource usage conditions is that aggregated usage for the at least one team has reached a predetermined threshold for a predetermined period of time and the at least one team is a TLB team.

10. The method of claim 4 wherein the at least one of the identified actionable resource usage conditions is that the at least one team is a NFT team having more than one standby network resource.

11. The method of claim 4 wherein the at least one of the identified actionable resource usage conditions is that the at least one team is a LB team having a team aggregate throughput less than a predetermined threshold for a predetermined period.

12. The method of claim 4 wherein the at least one of the identified actionable resource usage conditions is that the at least one team is a TLB team having a network resource over which data is not being balanced and is the recaptured network resource.

13. The method of claim 3 wherein:

the at least one team is a single network resource team;

the least one of the identified actionable resource usage conditions is that usage for the single network resource has reached a predetermined threshold for a predetermined period of time and the at least one team is a TLB team; and

said adding further comprises converting the at least one team from a single network resource team to a TLB team including the network resource allocated from the pool.

14. The method of claim 3 wherein:

the at least one team is a single network resource team;

the least one of the identified actionable resource usage conditions is that the single network resource has failed; and

said adding further comprises replacing the failed network resource with the network resource allocated from the pool.

15. The method of claim 3 wherein the least one of the identified actionable resource usage conditions is that the at least one team has less than two functional network resources;

and said adding further comprises replacing the failed network resource with the network resource from the pool.

16. The method of claim 5 wherein the least one of the identified actionable resource usage conditions is that the at least one team has less than two functional network resources;

and nonfunctional network resources of the at least one team are the network resources recaptured to the pool.

17. The method of claim 1 wherein said reconfiguring the network resources further comprises increasing transmit buffer resources for a network resource of at least one team when at least one of the identified actionable resource usage conditions is that the network resource is experiencing transmit underruns.

18. The method of claim 1 wherein said reconfiguring the network resources further comprises increasing receive buffer resources for a network resource of at least one team when at least one of the identified actionable resource usage conditions is that the network resource is experiencing receive underruns.

19. The method of claim 1 wherein said reconfiguring the network resources further comprises changing a LB algorithm for a LB team to a balance by IP address algorithm when at least one of the identified actionable resource usage conditions is that one NIC resource of the LB team is handling a majority of traffic transmitted by the LB team, most traffic transmitted by the team is destined for one MAC address, and traffic transmitted by the LB team is destined for multiple IP addresses.

20. The method of claim 4 wherein:

the at least one team is a TLB team;

at least one of the identified actionable resource usage conditions is that at least two members of the TLB team have different throughputs and two or more of the team members are running at a highest throughput level of the TLB team; and

said recapturing further comprises returning all team members running at a throughput less the than highest throughput for the team to the pool.

21. The method of claim 3 wherein:

the at least one team is a TLB team;

at least one of the identified actionable resource usage conditions is that at least two members of the TLB team have different throughputs and only one of the team members is running at a highest throughput level of the TLB team;

said reconfiguring further comprises changing the TLB team to a NFT team using the one member running at the highest throughput for the team as a primary and one of the other members as a standby; and

said recapturing further comprises returning the remaining members of the TLB team to the pool.

22. The method of claim 3 wherein:

the at least one team is a TLB team;

at least one of the identified actionable resource usage conditions is that at all members of the TLB team have the same throughputs and more than one member of the TLB team is receiving data having the TLB team'"'"'s MAC address; and

said reconfiguring further comprises changing the team from a TLB team to a SLB team.

23. The method of claim 3 wherein:

the at least one team is a NET team;

at least one of the identified actionable resource usage conditions is that at all members of the TLB team have the same throughputs; and

said reconfiguring further comprises changing the team from a NFT team to a TLB team.

24. The method of claim 4 wherein the at least one of the identified actionable resource usage conditions is that the pool is empty of functional network resources and said recapturing further comprises placing back in the pool a network resource form from one of the one more teams that is least adversely affected by said recapturing.

25. The method of claim 3 wherein:

the computer system is coupled to a network through a programmable switch; and

said reconfiguring further comprises programming the switch to couple the network resource to one or more network segments to which the at least one team is also coupled.

26. The method of claim 4 wherein:

the computer system is coupled to a network through a programmable switch; and

said recapturing further comprises programming the switch to decouple the network resource from one or more network segments to which the at least one team is also coupled.

27. The method of claim 1 wherein said initially allocating further comprises receiving inputs describing the initial allocation through a graphical user interface.

28. The method of claim 1 wherein said establishing one or more usage policies further comprises receiving inputs enabling one or more rules through a graphical user interface.

29. The method of claim 1 wherein at least one of the one or more established policies includes recapturing a network resource to the pool from a team of network resources when the team is detected as comprising underutilized network resources given the current usage of the team.

30. The method of claim 1 wherein at least one of the one or more established policies includes recapturing a network resource to the pool from a fault tolerance team when the team is detected as having an extra allocated network resource that is operable to provide fault tolerance.

31. The method of claim 1 wherein at least one of the one or more established policies includes recapturing a network resource to the pool from a NFT team when the team is detected as having more than one allocated network resource that is operable to provide fault tolerance.

32. The method of claim 1 wherein at least one of the one or more established policies includes recapturing a network resource to the pool from a load balancing team of network resources when usage of that network resource is determined to fall below a predetermined minimum level of usage for a predetermined duration of time.

33. The method of claim 1 wherein at least one of the one or more established policies includes recapturing a network resource to the pool from a load balancing team of network resources when usage of the team is determined to be not balanced across that network resource.

35. The method of claim 1 wherein at least one of the one or more established policies includes changing from a first algorithm by which traffic load is balanced over a load balancing team of network resources to a second load balancing algorithm when the second algorithm provides more optimal load balancing given current usage of the team.

36. The method of claim 1 wherein at least one of the one or more established policies includes changing the configuration of a team of resources from a fault tolerant team to a load balancing team when all network resources operate at a throughput that is substantially the same.

37. The method of claim 1 wherein at least one of the one or more established policies includes changing the configuration of a team of resources from a load balancing team to a fault tolerant team when at least one of the network resources operates at a disparate throughput.

38. The method of claim 1 wherein at least one of the one or more established policies includes adding at least one network resource from the pool to a load balancing team when the resources of the load balancing team have reached predetermined level of usage for a predetermined duration of time.

39. The method of claim 1 wherein at least one of the one or more established policies includes reprovisioning a network resource when it becomes low on communication resources.

40. The method of claim 1 wherein at least one of the one or more established policies includes adding at least one network resource from the pool to a team comprising a single network resource when the single network resource fails.

34. The method of claim I wherein at least one of the one or more established policies includes recapturing a network resource to the pool from a load balancing team of network resources when throughput of that network resource is substantially less than that of the other network resources of the load balancing team.

41. A computer system comprising:

a plurality of network resources each having a link for coupling processing resources of the computer system to at least one network device, each of the processing resources further having a driver associated therewith;

an intermediate teaming driver configured to run in conjunction with the computer system'"'"'s operating system, each instance of the driver operable to bind the drivers of two or more of the plurality of resources into a team;

a resource allocation application configured to run in conjunction with the operating system, the application operable to monitor the usage and status of network resources individually and the teams, to identify one or more actionable resource usage conditions based on usage policies defined by a set of extensible rules and to reconfigure the teams in accordance with the actionable resource usage conditions.

42. The computer system of claim 41 wherein the resource allocation application is configured to maintain a pool of resources from which unallocated resources are allocated to existing teams in accordance with the identified one or more actionable resource usage conditions.

43. The computer system of claim 42 wherein the resource allocation application is configured to recapture underutilized and failed network resources from the teams to the pool.

44. The computer system of claim 43 wherein the resource allocation application is configured to add a resource from the pool to a team by programming the at least one network device to couple the link associated with the added resource to network segments to which the team has been assigned.

45. The computer system of claim 44 wherein the resource allocation application is configured to recapture a resource from a team back to the pool by programming the at least one network device to decouple the link associated with the added resource from network segments to which the team has been assigned.

46. The computer system of claim 41 further comprising a configuration application, the configuration accessible through a graphical user interface by which the set of extensible rules are selected and parameterized.

47. A method of dynamically allocating and configuring network resources of a computer system, said method comprising:

initially allocating the network resources among one or more teams and a pool;

establishing one or more usage policies for at least one of the one or more teams;

continuously monitoring usage of the network resources to identify actionable resource usage conditions; and

reconfiguring the network resources in accordance with the one or more usage policies in response to the actionable resource usage conditions, said reconfiguring the network resources further comprising adding and recapturing network resources from the pool to at least one of the one or more teams in response to at least one of the identified actionable resource usage conditions.

48. The method of claim 47 wherein said establishing one or more usage policies further comprises receiving inputs enabling one or more rules through a graphical user interface.

49. A method of dynamically allocating and configuring network resources of a computer system, said method comprising:

initially allocating the network resources among one or more teams and a pool;

establishing one or more usage policies for at least one of the one or more teams, the usage policies established through selection of one or more rules through a graphical user interface;

continuously monitoring usage of the network resources to identify actionable resource usage conditions; and

reconfiguring the network resources in accordance with the one or more usage policies in response to the actionable resource usage conditions.

50. A computer system comprising a plurality of network resources, said computer system comprising:

means for initially allocating the network resources among one or more teams and a pool;

means for establishing one or more usage policies for at least one of the one or more teams;

means for continuously monitoring usage of the network resources to identify actionable resource usage conditions; and

means for reconfiguring the network resources in accordance with the one or more usage policies in response to the actionable resource usage conditions.

51. The computer system of claim 50 wherein said means for reconfiguring the network resources further comprises means for adding network resources from the pool to at least one of the one or more teams in response to at least one of the identified actionable resource usage conditions.

52. The computer system of claim 50 wherein said means for reconfiguring the network resources further comprises means for recapturing network resources from at least one of the one or more teams to the pool in response to at least one of the identified actionable resource usage conditions.

53. The computer system of claim 50 wherein said means for initially allocating further comprises means for receiving inputs describing the initial allocation through a graphical user interface.

54. The computer system of claim 50 wherein said means for establishing one or more usage policies further comprises means for receiving inputs enabling one or more rules through a graphical user interface.

55. The computer system of claim 51 wherein:

the computer system is coupled to a network through a programmable switch; and

said means for reconfiguring further comprises means for programming the switch to couple the network resource to one or more network segments to which the at least one team is also coupled.

56. The computer system of claim 52 wherein:

the computer system is coupled to a network through a programmable switch; and

said means for recapturing further comprises means for programming the switch to decouple the network resource from one or more network segments to which the at least one team is also coupled.