System and method for providing threshold-based access to compute resources

• US 8,370,898 B1
• Filed: 06/17/2005
• Issued: 02/05/2013
• Est. Priority Date: 06/18/2004
• Status: Active Grant

• Current Assignee: III Holdings 12, LLC (Intellectual Ventures LLC)
• Original Assignee: Adaptive Computing Enterprises, Inc. (Ala Services LLC)
• Inventors: Jackson, David Brian
• Primary Examiner(s): Zand, Kambiz
• Assistant Examiner(s): Sanders, Stephen
• Application Number:
  US11/155,347
• Time in Patent Office: 2,790 Days
• Field of Search: 726/2
• US Class Current: 726/2
• CPC Class Codes: G06F 2209/5014 Reservation
• CPC Class Codes: G06F 2209/504 Resource capping
• CPC Class Codes: G06F 9/4881 Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues
• CPC Class Codes: G06F 9/50 Allocation of resources, e.g. of the central processing unit [CPU]
• CPC Class Codes: G06F 9/5005 to service a request
• CPC Class Codes: G06F 9/5011 the resources being hardware resources other than CPUs, Servers and Terminals
• CPC Class Codes: G06F 9/5022 Mechanisms to release resources
• CPC Class Codes: G06F 9/5027 the resource being a machine, e.g. CPUs, Servers, Terminals
• CPC Class Codes: G06F 9/5072 Grid computing
• CPC Class Codes: H04L 41/00 Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
• CPC Class Codes: H04L 41/50 Network service management, e.g. ensuring proper service fulfilment according to agreements
• CPC Class Codes: Y02D 10/00 Energy efficient computing, e.g. low power processors, power management or thermal management

Abstract

The invention relates to systems, methods and computer-readable media for controlling access to compute resources in a compute environment such as a cluster or a grid. The method of providing conditional access to a compute environment comprises associating a required service level threshold with a compute environment, associating a service level with a requestor, receiving a request for access to the compute environment from the requestor; and, if the service level of the requestor meets the specified service level threshold, then allowing access to the compute resources. The threshold-based access may be enforced by reservations, policies or some other method.

Claims

1. A method comprising:

associating a service level threshold with a multi-node compute environment to yield a compute environment service level threshold, wherein each node in the multi-node compute environment has at least one processor and its own node-level operating system;

associating a service level with a human user;

receiving a request, at a first time, for access to the multi-node compute environment from the human user;

upon determining that the service level meets the compute environment service level threshold, allowing the human user access to the multi-node compute environment such that compute resources are reserved for use at a same time which is later than the first time; and

upon determining that the service level does not meet the compute environment service level threshold, denying the human user access to the multi-node compute environment.

2. The method of claim 1, further comprising:

enabling access to at least one compute resource within the multi-node compute environment for at least one specified timeframe, wherein access to the at least one compute resource is dependent on the service level satisfying the compute environment service level threshold.

3. The method of claim 2, wherein the compute environment service level threshold is associated with a timeframe through a use of a resource reservation and a resource reservation access control list.

4. The method of claim 1, wherein the compute environment service level threshold is associated with at least one of a job and a credential.

5. The method of claim 4, wherein the compute environment service level threshold relates to at least one of a current estimated queue time, current realized queue time, estimated minimum expansion factor, realized minimum expansion factor, estimated job throughput and realized job throughput.

6. The method of claim 4, wherein the compute environment service level threshold relates to at least one of a parameter that is met and a parameter that is exceeded.

7. The method of claim 6, wherein the compute environment service level threshold relates to fairshare usage of compute resources.

8. The method of claim 6, wherein the compute environment service level threshold relates to responsiveness for the credential.

9. The method of claim 7, wherein the fairshare usage relates to one of short term historical information and long term historical information.

10. The method of claim 8, wherein the compute environment service level threshold further relates to current compute resource utilization.

11. The method of claim 10, wherein the compute environment service level threshold further relates to current utilization associated with at least one of:

active jobs, allocated nodes, allocated processors, allocated processor seconds, bandwidth usage, licensing usage and disk storage usage.

12. The method of claim 1, wherein the compute environment service level threshold is associated with a usage parameter.

13. The method of claim 12, wherein the compute environment service level threshold is further associated with an average of resource usage.

14. The method of claim 13, wherein the average of the resource usage is associated with current job usage.

15. The method of claim 1, wherein receiving the request for access to the multi-node compute environment from the human user further comprises receiving a service request.

16. The method of claim 15, wherein the service request is for at least one of:

provisioning services, job migration services, data migration services, reservation creation, priority adjustment, policy exemption, limit exemption, preemption services, and access to other multi-node compute environment services.

17. The method of claim 16, wherein the other multi-node computer environment services comprise services provided by at least one of a database, a web server and an application server.

18. The method of claim 16, wherein the service request is related to a credential.

19. The method of claim 18, wherein the credential is at least one of:

a home cluster, a user, a group, an account, a class, a queue, a quality of service, and a service level agreement.

20. The method of claim 1, wherein the compute environment service level threshold relates to a measurable metric on at least one of a per job basis and a per credential basis.

21. The method of claim 1, wherein control of access to the multi-node compute environment relates to a credential.

22. The method of claim 21, wherein the credential is at least one of:

a home cluster, a user, a group, an account, a class, a queue, a quality of service and a service level agreement.

23. A non-transitory computer-readable medium storing instructions, the instructions comprising:

associating a service level threshold with a multi-node compute environment to yield a compute environment service level threshold, wherein each node in the multi-node compute environment has at least one processor and its own node-level operating system;

associating a service level with a human user;

receiving a request, at a first time, for access to the multi-node compute environment from the human user;

upon determining that the service level meets the compute environment service level threshold, allowing the human user access to the multi-node compute environment such that compute resources are reserved for use at a second time which is later than the first time; and

upon determining that the service level does not meet the compute environment service level threshold, denying the human user access to the multi-node compute environment.

24. A computing device comprising:

a processor; and

a computer-readable storage medium, which, when executed by the processor, control the processor to perform a method comprising;

associating a service level threshold with a multi-node compute environment to yield a compute environment service level, wherein each node in the multi-node compute environment has at least one processor and its own node-level operating system;

associating a service level with a human user;

receiving a request, at a first time, for access to the multi-node compute environment from the human user;

upon determining that the service level meets the compute environment service level threshold, allowing the human user access to the multi-node compute environment such that compute resources are reserved for use at a second time which is later than the first time; and

upon determining that the service level does not meet the compute environment service level threshold, denying the human user access to the multi-node compute environment.

25. A method comprising:

associating a service level threshold with a multi-node compute environment to yield a compute environment service level threshold, wherein each node in the multi-node compute environment has at least one processor and its own node-level operating system;

associating a service level with a requestor to yield a requestor service level, wherein the requestor is one of a human user and a group of human users;

receiving from the requestor at a first time a request for a reservation of compute resources in the multi-node compute environment, wherein the reservation is for use by a compute job at a second time which is later than the first time; and

if the requestor service level meets the compute environment service level threshold, then allowing the requestor access to the multi-node compute environment via the reservation.