Support of non-trivial scheduling policies along with topological properties

• US 10,467,051 B2
• Filed: 07/25/2016
• Issued: 11/05/2019
• Est. Priority Date: 05/11/2004
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Shpigelman, Igor
• Primary Examiner(s): Tang, Kenneth
• Application Number:
  US15/219,055
• Publication Number:
  US 20160335121A1
• Time in Patent Office: 1,198 Days
• Field of Search: None
• CPC Class Codes: G06F 13/1663 Access to shared memory
• CPC Class Codes: G06F 2209/483 Multiproc
• CPC Class Codes: G06F 2212/2542 Non-uniform memory access [NUMA] architecture
• CPC Class Codes: G06F 3/0619 in relation to data integrity, e.g. data losses, bit errors
• CPC Class Codes: G06F 3/0629 Configuration or reconfiguration of storage systems
• CPC Class Codes: G06F 3/0644 Management of space entities, e.g. partitions, extents, pools
• CPC Class Codes: G06F 3/0689 Disk arrays, e.g. RAID, JBOD
• 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 9/5016 the resource being the memory

Abstract

A system includes a scheduling unit for scheduling jobs to resources, and a library unit having a machine map of the system and a global status map of interconnections of resources. The library unit determines a free map of resources to execute the job to be scheduled, the free map indicating the interconnection of resources to which the job in a current scheduling cycle can be scheduled. A monitoring unit dispatches a job to the resources in the free map which match the resource mapping requirements of the job and fall within the free map.

Claims

1. A computer system, comprising:

a processor for executing computing instructions;

a memory for storing the computing instructions; and

a scheduling unit associated with the processor and the memory to schedule and dispatch a job having resource mapping requirements to resources in a computing architecture arranged at least in part on node boards in host computers via the executed computing instructions, the scheduling unit generating a candidate host list representing the resources in the host computers available to execute the job to be scheduled based on the resource mapping requirements of the job to be scheduled; and

a topology library unit including a machine map M of the computer system, the machine map M indicative of interconnections of the resources in the computer system to which the scheduling unit can schedule and dispatch the job, and at least one global status map Yn, each global status map Yn indicative of the interconnections of resources in the computer system to which the scheduling unit can schedule and dispatch the job if non-trivial scheduling requiring more than one scheduling cycle in which to schedule the job is utilized.

2. The computer system of claim 1, further comprising a topology monitoring unit for monitoring a status of the resources in the host computers and generating status information signals indicative of a status of the resources.

3. The computer system of claim 2, wherein the topology library unit receives the status information signals and the candidate host lists and determines a free map F of resources to execute the job to be scheduled, the free map indicative of the interconnection of resources to which the job in a current scheduling cycle can be scheduled based on the status information signals, the candidate host list and the machine map M.

4. The computer system of claim 3, wherein the scheduling unit also indicates resources which have a status other than free and to which the scheduling unit has determined the topology library unit may schedule the job being scheduled if non-trivial scheduling is utilized.

5. The computer system of claim 3, wherein the topology library unit initially determines a free map F of resources to execute the job being scheduled by removing from the machine map M all resources which fall within the at least one global status map Yn, and then re-introducing specific resources in the at least one global status map Yn which the scheduling unit has indicated the job being scheduled can be scheduled to.

6. A method, comprising:

receiving a job having resource mapping requirements, the job to be scheduled and dispatched to resources in a computing architecture arranged at least in part on node boards in host computers;

generating a candidate host list representing the resources in the host computers available to execute the job to be scheduled based on the resource mapping requirements of the job to be scheduled;

determining a machine map M of a computer system, the machine map M indicative of interconnections of the resources in the computer system to which a scheduling system can schedule the job;

determining at least one global status map Yn, each global status map Yn indicative of the interconnections of resources in the computer system to which the scheduling system can schedule the job if non-trivial scheduling requiring more than one scheduling cycle in which to schedule the job is utilized; and

scheduling and dispatching the job to the resources in the computer system based on the machine map M and the global status map Yn.

7. The method of claim 6, further comprising monitoring a status of the resources in the host computers and generating status information signals indicative of a status of the resources.

8. The method of claim 7, further comprising:

receiving the status information signals and the candidate host lists; and

determining a free map F of resources to execute the job to be scheduled, the free map indicative of the interconnection of resources to which the job in a current scheduling cycle can be scheduled based on the status information signals, the candidate host list and the machine map M.

9. The method of claim 8, further comprising indicating the resources which have a status other than free and to which the job may be scheduled if the non-trivial scheduling is utilized.

10. The method of claim 8, further comprising initially determining the free map F of the resources to execute the job being scheduled by removing from the machine map M resources which fall within the at least one global status map Yn, and then re-introducing specific resources in the at least one global status map Yn which the job being scheduled can be scheduled to.

11. A computer program product for scheduling jobs in a multiprocessor system, the computer program product comprising:

a non-transitory computer readable medium having computer readable program code embodied therewith, the computer readable program code comprising computer readable program code configured to;

receive a job having resource mapping requirements, the job to be scheduled and dispatched to resources in a computing architecture arranged at least in part on node boards in host computers;

generate a candidate host list representing the resources in the host computers available to execute the job to be scheduled based on the resource mapping requirements of the job to be scheduled;

determine a machine map M of a computer system, the machine map M indicative of interconnections of the resources in the computer system to which a scheduling system can schedule the job;

determine at least one global status map Yn, each global status map Yn indicative of the interconnections of resources in the computer system to which the scheduling system can schedule the job if non-trivial scheduling requiring more than one scheduling cycle in which to schedule the job is utilized; and

schedule and dispatch the job to the resources in the computer system based on the machine map M and the global status map Yn.

12. The computer program product of claim 11, wherein the computer readable program code is configured to monitor a status of the resources in the host computers and generate status information signals indicative of a status of the resources.

13. The computer program product of claim 12, wherein the computer readable program code is configured to:

receive the status information signals and the candidate host lists; and

determine a free map F of resources to execute the job to be scheduled, the free map indicative of the interconnection of resources to which the job in a current scheduling cycle can be scheduled based on the status information signals, the candidate host list and the machine map M.

14. The computer program product of claim 11, wherein the computer readable program code is configured to indicate the resources which have a status other than free and to which the job may be scheduled if the non-trivial scheduling is utilized.

15. The computer program product of claim 13, wherein the computer readable program code is configured to initially determine the free map F of the resources to execute the job being scheduled by removing from the machine map M resources which fall within the at least one global status map Yn, and then re-introduce specific resources in the at least one global status map Yn which the job being scheduled can be scheduled to.