Canceling and locking personal reservation if the workload associated with personal reservation exceeds window of time allocated within a resource reservation

• US 9,128,767 B2
• Filed: 10/06/2009
• Issued: 09/08/2015
• Est. Priority Date: 03/13/2004
• Status: Active Grant

• Current Assignee: III Holdings 12, LLC (Intellectual Ventures LLC)
• Original Assignee: Adaptive Computing Enterprises, Inc. (Ala Services LLC)
• Inventors: Jackson, David B.
• Primary Examiner(s): Kawsar, Abdullah Al
• Application Number:
  US12/573,967
• Publication Number:
  US 20100023949A1
• Time in Patent Office: 2,163 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06F 2209/5013 Request control
• CPC Class Codes: G06F 2209/5014 Reservation
• CPC Class Codes: G06F 2209/5021 Priority
• CPC Class Codes: G06F 2209/506 Constraint
• CPC Class Codes: G06F 2209/508 Monitor
• CPC Class Codes: G06F 9/5027 the resource being a machine, e.g. CPUs, Servers, Terminals
• CPC Class Codes: G06F 9/5038 considering the execution order of a plurality of tasks, e.g. taking priority or time dependency constraints into consideration scheduling strategies G06F9/4881 and subgroups
• CPC Class Codes: G06F 9/5061 Partitioning or combining of resources
• CPC Class Codes: G06F 9/5072 Grid computing
• CPC Class Codes: G06Q 40/00 Finance; Insurance; Tax strategies; Processing of corporate or income taxes
• CPC Class Codes: H04L 41/00 Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks

Abstract

A system and method are disclosed for dynamically reserving resources within a cluster environment. The method embodiment of the invention comprises receiving a request for resources in the cluster environment, monitoring events after receiving the request for resources and based on the monitored events, dynamically modifying at least one of the request for resources and the cluster environment.

Claims

1. A method comprising:

reserving resources in a compute environment under a first reservation for consumption by a workload to yield reserved resources, wherein the compute environment comprises a plurality of network nodes and wherein the first reservation of resources is in a window of time in which a user submits personal reservations;

receiving a personal reservation for the use of the reserved resources within the window of time;

providing access to the reserved resources for the personal reservation to process the workload;

monitoring events associated with the workload consuming the reserved resources to yield a monitored event; and

based on the monitored event and after the workload begins to consume the reserved resources, dynamically modifying the first reservation to establish a second reservation such that the workload, before completing, consumes different resources than would have been consumed under the first reservation, wherein if a received consumption workload associated with the personal reservation will exceed the window of time for the first reservation of resources, then the method comprises canceling and locking out the personal reservation from access to the reserved resources.

2. The method of claim 1, wherein the compute environment is one of a compute farm, a cluster environment and a grid environment.

3. The method of claim 1, wherein the first reservation and the second reservation are associated with one of consumption resources, provisioning resources, direct volume access, and batch workload processing.

4. The method of claim 1, wherein the first reservation and the second reservation establish a virtual private cluster.

5. The method of claim 1, wherein monitoring events after the first time further comprises monitoring the compute environment via a common administrative control.

6. The method of claim 1, wherein monitoring events after the first time further comprises monitoring to determine if a party has submitted the workload for processing under the first reservation.

7. The method of claim 6, wherein if the party has not submitted the workload for processing after a predetermined amount of time after the first time, then dynamically modifying the first reservation further comprises canceling the first reservation.

8. The method of claim 7, wherein the workload comprises one of a reservation, an object that monitors policy, an object that monitors credentials, an object that monitors node states via a common administrative control and an object that monitors the compute environment via the common administrative control.

9. The method of claim 8, wherein modifying the compute environment based on the monitored events further comprises dynamically modifying the compute environment to satisfy a request for resources associated with the first reservation.

10. The method of claim 9, wherein dynamically modifying the compute environment further comprises one of:

modifying a node, modifying an operating system, installing end user applications, dynamically partitioning node resources and adjusting network configuration.

11. The method of claim 1, wherein monitoring events further comprises monitoring resources associated with the first reservation.

12. The method of claim 1, wherein monitoring events further comprises monitoring the workload submitted by a user.

13. The method of claim 12, wherein if the workload submitted within the first reservation will extend beyond the first reservation, the method further comprises canceling the workload.

14. The method of claim 13, wherein prior to canceling the workload, the method further comprises presenting to the user that submitted the workload an option of extending the first reservation to accommodate the workload.

15. The method of claim 14, wherein the option of extending the first reservation to accommodate the workload is subject to pre-established policies.

16. The method of claim 15, further comprising presenting to the user with an option of extending the first reservation and a pricing option to extend the first reservation.

17. The method of claim 1, wherein personal reservations are one of a non-administrator reservation and an administrator reservation.

18. The method of claim 1, wherein the first reservation for a window of time is a request for resources for a periodic window of time.

19. The method of claim 18, wherein the periodic window of time is daily, weekly, monthly, quarterly or yearly.

20. The method of claim 1, wherein if a received consumption workload associated with the personal reservation will exceed the window of time, then the method comprises never starting the consumption workload.

21. The method of claim 1, further comprising, before canceling and locking out the personal reservation, the step of:

presenting to a user who submitted the personal reservation an option of allowing the workload running within the personal reservation to complete although a time for completing the remaining workload is beyond the window of time for the personal reservation.

22. The method of claim 1, further comprising, if the workload submitted under a personal reservation would exceed the personal reservation, extending the personal reservation to meet the needs of the workload.

23. A method of dynamically modifying a reservation of resources within a compute environment comprising a plurality of compute nodes under common administrative control, the method comprising:

receiving a request for a reservation for resources in the compute environment, wherein the request for resources in the compute environment comprises a request for a reservation of resources for a window of time in which at least one user can submit personal reservations;

based on the request, reserving a set of resources in the compute environment to yield reserved resources;

receiving a personal reservation for the use of resources within the window of time to yield personal reserved resources;

providing access to the personal reserved resources to process workload;

monitoring events during which the workload is scheduled to begin to consume the personal reserved resources; and

based on the events and after the workload begins consuming the personal reserved resources, dynamically modifying the reservation to create a modified reservation in which a different set of resources, relative to the set of resources, is reserved under the modified reservation, wherein the workload, prior to completing, consumes the different set of resources, wherein if a received consumption workload associated with the personal reservation will exceed the window of time for the reservation of resources, then the method further comprises canceling and locking out the personal reservation from access to the reserved resources.

24. The method of claim 23, wherein dynamically modifying the reservation comprises migrating the reservation to be associated with the different set of resources.

25. The method of claim 24, wherein migrating the reservation is one of a migration in space and a migration in time to the different set of resources.

26. The method of claim 25, wherein the migration is the migration in time and wherein the migration in time creates the modified reservation at an earliest time possible.

27. The method of claim 25, wherein the migration in time creates the modified reservation based on availability of resources in the compute environment.

28. The method of claim 25, wherein the migration is a migration in space, wherein the migration comprises migrating the reservation to the different set of resources that will provide better performance of the compute environment for the request for resources relative to the set of resources.

29. The method of claim 25, wherein the migration is a migration in space and wherein the migration in space comprises migrating the reservation to resources according to a failure or projected failure of resources.

30. The method of claim 23, wherein the modified reservation comprises modifying a time associated with the reservation of the set of resources.

31. The method of claim 23, wherein the different set of resources better meet needs associated with the request for resources relative to the set of resources.

32. The method of claim 23, wherein monitoring events after receiving the request for resources further comprises monitoring workload submitted within a reservation based on the request.

33. The method of claim 32, wherein if the workload submitted within the reservation will extend beyond the reservation, the method further comprises canceling the workload.

34. The method of claim 33, wherein prior to canceling the workload, the method further comprises presenting to an entity that submitted the workload an option of modifying the reservation to accommodate the workload.

35. The method of claim 34, wherein the option of modifying the reservation to accommodate the workload is subject to pre-established policies.

36. The method of claim 35, further comprising presenting an entity with an option to extend the reservation and a pricing option to extend the reservation.

37. The method of claim 23, wherein personal reservations are one of a non-administrator reservation and an administrator reservation.

38. The method of claim 23, wherein the reservation of resources for a window of time is a request for cluster resources for a periodic window of time.

39. The method of claim 38, wherein the periodic window of time may be daily, weekly, monthly, quarterly or yearly.

40. The method of claim 23, wherein if the received consumption workload associated with the personal reservation will exceed the window of time, then the method comprises never starting the received consumption workload.

41. The method of claim 23, further comprising, before canceling and locking out the personal reservation, the step of:

presenting to a user who requested the personal reservation an option of allowing the workload running within the personal reservation to complete although it is beyond the window of time for the reserved resources.

42. The method of claim 41, further comprising, if the workload submitted under a personal reservation would exceed the personal reservation, extending the personal reservation to meet the needs of the workload.

43. A non-transitory computer-readable medium storing instructions for controlling a computing device to dynamically manage resources within a compute environment comprising a plurality of compute nodes under common administrative control, the instructions causing the computing device to perform operations comprising:

receiving a request for a reservation of resources in the compute environment, wherein the request for resources in the compute environment comprises a request for a reservation of resources for a window of time in which at least one user can submit personal reservations;

based on the request, reserving a set of resources in the compute environment to yield reserved resources;

receiving a personal reservation for the use of resources within the window of time to yield personal reserved resources for the personal reservation;

providing access to the personal reserved resources for the personal reservation to process workload;

monitoring events during which the workload is consuming the personal reserved resources; and

based on the monitored events, dynamically modifying the reservation to create a modified reservation in which a different set of resources, relative to the set of resources, is reserved under the modified reservation, wherein the workload, prior to completing, consumes the different set of resources, wherein if a received consumption workload associated with the personal reservation will exceed the window of time for the reservation of resources, then the method further comprises canceling and locking out the personal reservation from access to the reserved resources.

44. A system for dynamically managing resources within a compute environment comprising a plurality of compute nodes under common administrative control, the system comprising:

a processor; and

a computer-readable medium storing instructions which, when executed by the processor, control the processor to perform operations comprising;

receiving a request for a reservation of resources in the compute environment, wherein the request for resources in the compute environment comprises a request for a reservation of resources for a window of time in which at least one user can submit personal reservations;

based on the request, reserving a set of resources in the compute environment to yield reserved resources;

receiving a personal reservation for the use of resources within the window of time to yield personal reserved resources;

providing access to the personal reserved resources to process workload;

monitoring events during which the workload is consuming the personal reserved resources; and

based on the monitored events, dynamically modifying the reservation to create a modified reservation in which a different set of resources, relative to the set of resources, is reserved under the modified reservation, wherein the workload, prior to completing, consumes the different set of resources, wherein if a received consumption workload associated with the personal reservation will exceed the window of time for the reservation of resources, then the method further comprises canceling and locking out the personal reservation from access to the reserved resources.

45. A compute environment comprising:

a plurality of computing devices having a plurality of processors and the plurality of computing devices are under common administrative control, the compute environment having resources which are dynamically managed according to operations comprising;

receiving a request from a user for a reservation of resources in the compute environment, wherein the request for resources in a compute environment comprises a request for a reservation of resources for a window of time in which at least one user can submit personal reservations;

based on the request, reserving a set of resources in the compute environment to yield reserved resources;

receiving a personal reservation for the use of resources within the window of time to yield personal reserved resources;

providing access to the personal reserved resources to process workload;

monitoring events during which the workload is consuming the personal reserved resources; and

based on the monitored events, dynamically modifying the reservation to create a modified reservation in which a different set of resources, relative to the set of resources, is reserved under the modified reservation, wherein the workload, prior to completing, consumes the different set of resources, wherein if a received consumption workload associated with the personal reservation will exceed the window of time for the reservation of resources, then the method further comprises canceling and locking out the personal reservation from access to the reserved resources.