Systems and methods for providing dynamic quality of service for a distributed system

• US 20040030777A1
• Filed: 03/19/2003
• Published: 02/12/2004
• Est. Priority Date: 09/07/2001
• Status: Active Grant

• Current Assignee: Oracle America, Inc. (Oracle Corporation)
• Original Assignee: Oracle America, Inc. (Oracle Corporation)
• Inventors: Mitchell, Larry J., Reedy, Dennis G.
• Granted Patent: US 7,660,887 B2
• Time in Patent Office: Days
• US Class Current: 709/224
• CPC Class Codes: G06F 11/3409 for performance assessment
• CPC Class Codes: G06F 11/3442 for planning or managing the needed capacity
• CPC Class Codes: G06F 2201/81 Threshold
• CPC Class Codes: G06F 2201/86 Event-based monitoring
• CPC Class Codes: G06F 9/542 Event management; Broadcasting; Multicasting; Notifications
• CPC Class Codes: H04L 41/12 Discovery or management of network topologies
• CPC Class Codes: H04L 41/5051 Service on demand, e.g. definition and deployment of services in real time
• CPC Class Codes: H04L 43/00 Arrangements for monitoring or testing data switching networks
• CPC Class Codes: H04L 43/16 Threshold monitoring
• CPC Class Codes: H04L 67/10 in which an application is distributed across nodes in the network software deployment G06F8/60; multiprogramming arrangements G06F9/46
• CPC Class Codes: H04L 67/51 Discovery or management thereof, e.g. service location protocol [SLP] or web services
• CPC Class Codes: H04L 69/329 in the application layer [OSI layer 7]
• CPC Class Codes: H04L 9/40 Network security protocols

Abstract

Methods and systems consistent with certain aspects related to the present invention provide a dynamic service delivery system based on a capability to provide services through a distributed system using qualitative and quantitative Quality of Service (QoS) attributes. In accordance with one aspect of the invention, a method for providing a service that includes collecting capability information reflecting a current operating condition of a compute resource and determining a platform capability and a measurable capability value for a depletion-oriented characteristic of the compute resource based on the collected capability information. Based on at least one of the platform capability and the measurable capability value; a capability value for the compute resource is determined and this value is provided to a provision manager that determines whether the compute resource is worthy of being an available resource in the distributed system based on the capability value. Additionally, the provision manager may determine which of a plurality of services provided by compute resources is more capable of handling a service request based on service capability values corresponding to platform capabilities and/or the depletion-oriented characteristics of each of the compute resources.

Claims

1. A method for providing a service in a distributed system, comprising:

collecting capability information reflecting a current operating condition of a compute resource;

determining a measurable capability value for a depletion-oriented characteristic of the compute resource based on the collected capability information;

determining a platform capability for the compute resource;

determining a capability value for the compute resource based on at least one of the platform capability and the measurable capability value; and

providing the capability value to a provision manager that is capable of determining whether the compute resource is to be designated an available resource in the distributed system based on the capability value.

2. The method of claim 1, wherein the depletion-oriented characteristic is one of memory usage, processor utilization, network bandwidth usage, and a number of executing operating system resources.

3. The method of claim 1, wherein the platform capability reflects at least one of hardware and software based capabilities that are associated with in the compute resource.

4. The method of claim 1, wherein the capability value reflects an overall ability for the compute resource to provide a service based on at least one of the platform capability and the measurable capability value.

5. The method of claim 4, wherein the overall ability for the compute resource reflected in the capability value is directly proportional to a fluctuation of the depletion-oriented characteristic.

6. The method of claim 1, wherein the provision manager determines whether the compute resource is to be designated an available resource in the distributed system based on the capability value and another capability value associated with another compute resource included in the distributed system.

7. The method of claim 1, further comprising:

providing a service to a requesting entity based on a determination from the provision manager that the compute resource is more capable of providing the service than another compute resource included in the distributed system.

8. The method of claim 7, wherein the determination from the provision manager that the compute resource is more capable of providing the service than the another compute resource is based on the capability value.

9. The method of claim 1, wherein providing the capability value includes:

creating a resource capability object that includes the capability value; and

providing the resource capability object to the provision manager.

10. The method of claim 9, wherein the compute resource may update the resource capability object based on changes to at least one of the platform capability and the depletion-oriented characteristic.

11. The method of claim 10, wherein the compute resource updates the resource capability object by invoking a method provided by the provision manager.

12. The method of claim 1, wherein the capability information is collected periodically.

13. The method of claim 1, wherein the capability value is provided to the provision manager periodically.

59. A computer-readable medium including instructions for, when executed by a processor, performing the method of claim 1.

14. A method for providing a service in a distributed system, comprising:

collecting a resource capability object from a compute resource;

determining whether the compute resource satisfies a QoS capability threshold based on the resource capability object; and

registering the computer resource as an available resource in the distributed system based on the determination.

15. The method of claim 14, wherein the QoS capability is one of a platform capability and a measurable capability.

16. The method of claim 15, wherein the platform capability is associated with a capability provided by a hardware or software mechanism operating within the compute resource.

17. The method of claim 15, wherein the measurable capability is associated with a depletion-oriented characteristic of the compute resource.

18. The method of claim 17, wherein the depletion-oriented characteristic is one of a memory usage, processor utilization, network bandwidth usage, and a number of executing operating system resources associated with the compute resource.

19. The method of claim 14, wherein determining whether the compute resource satisfies a QoS capability threshold includes:

determining whether a resource capability value reflecting a capability associated with a depletion-oriented characteristic of the compute resource is one of greater than, equal to, and less than, a predetermined capability threshold value.

20. The method of claim 19, wherein the resource capability value is determined by the compute resource and is included in the resource capability object provided by the compute resource.

21. The method of claim 14, wherein registering the computer resource as an available resource includes:

providing an entry for the compute resource in a data structure including other identifiers for other registered compute resources.

22. The method of claim 14, wherein the compute resource provides a first service and the method further comprising:

receiving a service request from a requesting entity; and

determining that the compute resource is more capable of providing the requested service to the requesting entity using the first service than a second compute resource included in the distributed system that provides a second service.

23. The method of claim 22, wherein determining that the compute resource is more capable of providing the service includes:

comparing a first cost value associated with the first service and a second cost value associated with the second service; and

providing the first service to the requesting entity based on the comparison.

24. The method of claim 23, wherein the first and second cost values are associated with one or more depletion-oriented characteristics of the compute and second compute resources, respectively.

25. The method of claim 14, wherein determining whether the compute resource satisfies a QoS capability threshold includes:

comparing a resource capability value associated with a depletion-oriented characteristic of the compute resource with a corresponding QoS capability threshold value.

26. A method of providing a service in a distributed system including a cybernode running on a compute resource that provides the service, the method comprising:

monitoring a depletion-oriented characteristic of the compute resource;

determining that a QoS capability of the compute resource does not satisfy a corresponding QoS threshold;

providing, based on the determination, to an SLA manager a threshold event that includes an indication of the QoS capability; and

performing an action based on the threshold event.

27. The method of claim 26, further comprising:

registering the SLA manager with a QoS framework within the cybernode to receive the threshold event.

28. The method of claim 27, wherein the QoS framework includes a monitoring object that monitors one or more depletion-oriented characteristics of the compute resource.

29. The method of claim 28, wherein the monitoring object is contained within a measurable capability object that provides the threshold event to the SLA manager.

30. The method of claim 26, wherein the corresponding SLA threshold value reflects an acceptable level of capability associated with the QoS capability.

31. The method of claim 26, wherein the QoS capability is one of a depletion-oriented characteristic and a platform capability of the compute resource.

32. The method of claim 31, wherein the depletion-oriented characteristic is one of processing component utilization and memory usage.

33. The method of claim 31, wherein the platform capability is a capability provided by one or more hardware or software based mechanism operating within the compute resource.

34. The method of claim 26, wherein providing the threshold event to the SLA manager further includes:

providing an SLA threshold event from the SLA manager to an enforcement module that includes one or more rules that are used to determine the action.

35. The method of claim 34, wherein the enforcement module is included in a framework associated with the cybernode.

36. The method of claim 26, wherein performing the action includes:

determining the action based on the QoS capability that does not satisfy the SLA threshold value.

37. The method of claim 26, wherein the action includes canceling a lease held by the cybernode with a provision manager that provides service requests to the compute resource based on a resource capability value associated with the QoS capability.

38. The method of claim 26, wherein the action includes one of canceling a lease held by the cybernode with a provision manager, providing a notification to a user identifying the depletion-oriented characteristic, and moving the service provided by the compute resource to another compute resource included in the distributed system.

39. The method of claim 26, wherein the QoS capability is a depletion-oriented characteristic of the compute resource and the QoS capability threshold is a value associated with the depletion-oriented characteristic defined within a SLA.

40. The method of claim 26, wherein determining that a QoS capability of the compute resource does not meet a corresponding QoS threshold includes:

determining that a service configuration count of the compute resource exceeds a predetermined value.

41. The method of claim 26, wherein determining that a QoS capability of the compute resource does not satisfy a corresponding QoS threshold includes:

comparing a resource capability value associated with a depletion-oriented characteristic of the compute resource with a corresponding QoS capability threshold value.

42. A system for providing a service in a distributed system, comprising:

a provision manager configured to provide a service to a requesting entity based on a service capability value corresponding to each of a plurality of services instantiated in the distributed system; and

a compute resource that provides one of the instantiated services configured to monitor a QoS capability of the compute resource, determine a resource capability value of the compute resource reflecting an overall ability for the compute resource to provide the instantiated service based on the monitored QoS capability, and provide the capability value to the provision manager.

43. The system of claim 42, wherein the provision manager is further configured to determine whether the compute resource is to be designated an available resource in the distributed system based on the capability value.

44. The system of claim 42, wherein the provision manager determines that the compute resource is capable of providing the instantiated service to the requesting entity based on a service capability value associated with the instantiated service.

45. The system of claim 44, wherein the service capability value associated with the instantiated service is based on the resource capability value.

46. The system of claim 43, wherein the compute resource is configured to remove itself as an available resource based on a determination that the QoS capability does not meet a QoS threshold.

47. The system of claim 42, further including:

an enforcement module that performs an action based on an event provided by the compute resource indicating that the QoS capability does not meet a QoS threshold.

48. The system of claim 47, wherein the action includes canceling a lease held by the compute resource with the provision manager.

49. The system of claim 47, wherein the action includes one of canceling a lease held by the compute resource with a provision manager, providing a notification to a user identifying the depletion-oriented characteristic, and moving the service provided by the compute resource to another compute resource included in the distributed system.

50. The system of claim 42, wherein the compute resource includes a monitoring object that periodically monitors the QoS capability of the compute resource.

51. The system of claim 42, wherein the compute resource includes a compute resource object that determines the resource capability value of the compute resource.

52. The system of claim 51, wherein the compute resource object provides a resource capability object including the capability value to the provision manager.

53. The system of claim 42, wherein the resource capability value is a value that is independent of a type of platform of the compute resource.

54. The system of claim 43, wherein the provision manager compares the resource capability value of the compute resource with a second resource capability value of a second compute resource to determine whether the compute resource is worthy of being an available resource in the distributed system.

55. The system of claim 54, wherein the second compute resource includes a platform architecture that is different that a platform architecture of the compute resource.

56. A system for providing a service in a distributed system, comprising:

means for collecting capability information reflecting a current operating condition of a compute resource;

means for determining a measurable capability value for a depletion-oriented characteristic of the compute resource based on the collected capability information;

means for determining a platform capability for the compute resource;

means for determining a capability value for the compute resource based on at least one of the platform capability and the measurable capability value; and

means for providing the capability value to a provision manager that is capable of determining whether the compute resource is to be designated an available resource in the distributed system based on the capability value.

57. A system for providing a service in a distributed system, comprising:

means for collecting a resource capability object from a compute resource;

means for determining whether the compute resource satisfies a QoS capability threshold based on the resource capability object; and

means for registering the computer resource as an available resource in the distributed system based on the determination.

58. A system for providing a service in a distributed system including a cybernode running on a compute resource that provides the service, the method comprising:

means for monitoring a depletion-oriented characteristic of the compute resource;

means for determining that a QoS capability of the compute resource does not satisfy a corresponding QoS threshold;

means for providing to an SLA manager a threshold event that includes an indication of the QoS capability; and

means for performing an action based on the threshold event.

60. A computer-readable medium including instructions for performing a method, when executed by a processor, for providing a service in a distributed system, the method comprising:

collecting a resource capability object from a compute resource;

determining whether the compute resource satisfies a QoS capability threshold based on the resource capability object; and

registering the computer resource as an available resource in the distributed system based on the determination.

61. A computer-readable medium including instructions for performing a method, when executed by a processor, for providing a service in a distributed system including a cybernode running on a compute resource that provides the service, the method comprising:

monitoring a depletion-oriented characteristic of the compute resource;

determining that a QoS capability of the compute resource does not satisfy a corresponding QoS threshold;

providing to an SLA manager a threshold event that includes an indication of the QoS capability; and

performing an action based on the threshold event.

62. A system for providing a service, comprising:

a memory including;

a set of instructions for receiving a resource capability value from each of a number of compute resources,

a set of instructions for determining which of the compute resources is more capable of providing a service to a requesting entity based on each compute resource'"'"'s capability value, and

a set of instructions for enabling one of the compute resources to provision the service to the requesting entity based on the determination; and

a processor for executing the sets of instructions included in the memory.

63. A system comprising:

a memory including;

a set of instructions for determining a resource capability value representing an overall ability of the compute resource to provision a service,

a set of instructions for registering as an available resource with a provision manager based on the resource capability value, and

a set of instructions for monitoring QoS capabilities of the compute resource to determine whether the compute resource is currently able to provision the service based on one or more QoS threshold values; and

a processor for executing the sets of instructions included in the memory.

64. A Service Level Agreement (SLA) manager included in a cybernode of a compute resource that is registered with a provision manager as an available resource based on a universal capability value reflecting an overall ability to provision a service, the SLA manager executable by a processor, comprising:

program code for registering with a measurable capability object to receive a threshold event based on a determination that the compute resource has a QoS capability that does not meet a predetermined QoS threshold; and

program code for providing an SLA event message to an enforcement module that determines an action to be performed based on the QoS capability that does not meet the predetermined QoS threshold.