Pessimistic scheduling for topology optimized workload placement

Patent metadata

Field	Value
Publication number	US-10979531-B2
Application number	US-201916264570-A
Country	US
Kind code	B2
Filing date	Jan 31, 2019
Priority date	Jan 31, 2019
Publication date	Apr 13, 2021
Grant date	Apr 13, 2021

Abstract

Various embodiments are provided for using pessimistic scheduling for topology optimized workload placement by a processor in a computing environment. An excessive amount of computing resources may be requested (e.g., a pessimistic request) to execute a workload as compared to a required amount of the computing resources to execute the workload. The workload may be assigned to a selected configuration of the excessive amount of computing resources and releasing a remaining amount of the excessive amount of computing resources.

First claim

The invention claimed is:

1. A method for using pessimistic scheduling for topology optimized workload placement by a processor in a computing environment, comprising:
requesting, by a resource requester to a workload scheduler, an excessive amount of computing resources to execute a workload, the excessive amount of computing resources known to the resource requester prior to the request to be more resources than necessary to execute the workload as compared to a required amount of the computing resources necessary to execute the workload, wherein a number of the excessive amount of computing resources able to be requested by the resource requester is limited to a predefined percentage over the required amount of the computing resources necessary to execute the workload for a given number of subnets connecting the computing resources; and
assigning, by the workload scheduler, the workload to a selected configuration of the excessive amount of computing resources and releasing a remaining amount of the excessive amount of computing resources over which are more resources than necessary to execute the workload.

2. The method of claim 1 , further including performing one or more micro-benchmark operations to simulate behavior of the workload.

3. The method of claim 1 , further including estimating workload behavior of the workload according to one or more various configuration of the excessive amount of computing resources.

4. The method of claim 1 , further including identifying a subset of the excessive amount of computing resources that maximizes performance of the workload.

5. The method of claim 1 , further including ranking each of a plurality of configuration of the excessive amount of computing resources according to an estimated workload performance.

6. The method of claim 5 , further including selecting, as the selected configuration of the excessive amount of computing resources, one of the plurality of configuration of the excessive amount of computing resources according to the ranking.

7. The method of claim 1 , further including:
retaining the remaining amount of the excessive amount of computing resources for a selected time period; and
releasing the remaining amount of the excessive amount of computing resources back to a cluster of computing resources upon expiration of the selected time period.

8. A system for using pessimistic scheduling for topology optimized workload placement by a processor in a computing environment, comprising:
one or more computers with executable instructions that when executed cause the system to:
request, by a resource requester to a workload scheduler, an excessive amount of computing resources to execute a workload, the excessive amount of computing resources known to the resource requester prior to the request to be more resources than necessary to execute the workload as compared to a required amount of the computing resources necessary to execute the workload, wherein a number of the excessive amount of computing resources able to be requested by the resource requester is limited to a predefined percentage over the required amount of the computing resources necessary to execute the workload for a given number of subnets connecting the computing resources; and
assign, by the workload scheduler, the workload to a selected configuration of the excessive amount of computing resources and releasing a remaining amount of the excessive amount of computing resources over which are more resources than necessary to execute the workload.

9. The system of claim 8 , wherein the executable instructions further perform one or more micro-benchmark operations to simulate behavior of the workload.

10. The system of claim 8 , wherein the executable instructions further estimate workload behavior of the workload according to one or more various configuration of the excessive amount of computing resources.

11. The system of claim 8 , wherein the executable instructions further identify a subset of the excessive amount of computing resources that maximizes performance of the workload.

12. The system of claim 8 , wherein the executable instructions further rank each of a plurality of configuration of the excessive amount of computing resources according to an estimated workload performance.

13. The system of claim 12 , wherein the executable instructions further select, as the selected configuration of the excessive amount of computing resources, one of the plurality of configuration of the excessive amount of computing resources according to the ranking.

14. The system of claim 8 , wherein the executable instructions further:
retain the remaining amount of the excessive amount of computing resources for a selected time period; and
release the remaining amount of the excessive amount of computing resources back to a cluster of computing resources upon expiration of the selected time period.

15. A computer program product for using pessimistic scheduling for topology optimized workload placement by a processor in a computing environment, the computer program product comprising a non-transitory computer-readable storage medium having computer-readable program code portions stored therein, the computer-readable program code portions comprising:
an executable portion that requests, by a resource requester to a workload scheduler, an excessive amount of computing resources to execute a workload, the excessive amount of computing resources known to the resource requester prior to the request to be more resources than necessary to execute the workload as compared to a required amount of the computing resources necessary to execute the workload, wherein a number of the excessive amount of computing resources able to be requested by the resource requester is limited to a predefined percentage over the required amount of the computing resources necessary to execute the workload for a given number of subnets connecting the computing resources; and
an executable portion that assigns, by the workload scheduler, the workload to a selected configuration of the excessive amount of computing resources and releasing a remaining amount of the excessive amount of computing resources over which are more resources than necessary to execute the workload.

16. The computer program product of claim 15 , further including an executable portion that performs one or more micro-benchmark operations to simulate behavior of the workload.

17. The computer program product of claim 15 , further including an executable portion that estimates workload behavior of the workload according to one or more various configuration of the excessive amount of computing resources.

18. The computer program product of claim 15 , further including an executable portion that identifies a subset of the excessive amount of computing resources that maximizes performance of the workload.

19. The computer program product of claim 15 , further including an executable portion that:
ranks each of a plurality of configuration of the excessive amount of computing resources according to an estimated workload performance; and
selects, as the selected configuration of the excessive amount of computing resources, one of the plurality of configuration of the excessive amount of computing resources according to the ranking.

20. The computer program product of claim 15 , further including an executable portion that:
retains the remaining amount of the excessive amount of computing resources for a selected time period; and
releases the remaining amount o

Assignees

• IBM

Inventors

• Seelam Seetharami
• Cho Minsik

Classifications

• H04L41/12

  Discovery or management of network topologies · CPC title

• H04L47/83

  based on usage prediction · CPC title

• H04L41/0897

  by horizontal or vertical scaling of resources, or by migrating entities, e.g. virtual resources or entities · CPC title

• H04L41/40

  using virtualisation of network functions or resources, e.g. SDN or NFV entities · CPC title

• H04L67/60

  Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources (admission control or resource allocation H04L47/70) · CPC title