Resource adaptation using nonlinear relationship between system performance metric and resource usage

Patent metadata

Field	Value
Publication number	US-11868810-B2
Application number	US-202016741962-A
Country	US
Kind code	B2
Filing date	Jan 14, 2020
Priority date	Nov 15, 2019
Publication date	Jan 9, 2024
Grant date	Jan 9, 2024

Abstract

Techniques are provided for allocating resources for one or more workloads. One method comprises obtaining a current performance of a workload; determining an adjustment to a current allocation of a resource allocated to the workload by evaluating a representation of a relationship between: (i) the current allocation of the resource allocated to the workload, (ii) a performance metric, and (iii) the current performance of the workload; and initiating an application of the determined adjustment to the current allocation of the resource for the workload. The performance metric may comprise a nominal value of a predefined service metric and the current performance of the workload may comprise a current value of a variable that tracks a given predefined service metric of the workload. An amount (or percentage) of the adjustment permitted for each iteration may be controlled. A sum of allocated resources can be constrained to an amount of available resources.

First claim

What is claimed is:

1. A method, comprising:
obtaining a current performance of at least one iterative workload;
determining an adjustment to a current allocation of at least one resource allocated to the at least one iterative workload by evaluating, for each iteration of the at least one iterative workload, a representation of a relationship between: (i) the current allocation of the at least one resource allocated to the at least one iterative workload, (ii) a performance metric, wherein a value of the performance metric is changed for at least one iteration of the at least one iterative workload and is maintained at least for the at least one iteration, and (iii) the current performance of the at least one iterative workload, wherein the determined adjustment to the current allocation of the at least one resource allocated to the at least one iterative workload is controlled to be between a first upper limit value and a second independent lower limit value and wherein one or more of the first upper limit value and the second independent lower limit value is: (i) changed for a plurality of iterations of the at least one iterative workload and (ii) determined using a percentage of the current allocation of the at least one resource; and
initiating an application of the determined adjustment to the current allocation of the at least one resource for the at least one iterative workload,
wherein the method is performed by at least one processing device comprising a processor coupled to a memory.

2. The method of claim 1 , wherein the performance metric comprises a nominal value of a predefined service metric.

3. The method of claim 1 , wherein the current performance of the at least one iterative workload comprises a current value of a variable that tracks a given predefined service metric of the at least one iterative workload.

4. The method of claim 3 , wherein a current error of the variable that tracks the given predefined service metric of the at least one iterative workload comprises a difference between the current value of the given predefined service metric and a corresponding predefined target value for the given predefined service metric.

5. The method of claim 1 , wherein the at least one iterative workload comprises one workload and wherein the representation comprises an analytic representation of a quadratic representation of the relationship.

6. The method of claim 1 , wherein the at least one iterative workload comprises a plurality of workloads and wherein the representation comprises a quadratic representation of the relationship.

7. The method of claim 1 , wherein a sum of the at least one resource is constrained to an available amount of the at least one resource.

8. An apparatus comprising:
at least one processing device comprising a processor coupled to a memory;
the at least one processing device being configured to implement the following steps:
obtaining a current performance of at least one iterative workload;
determining an adjustment to a current allocation of at least one resource allocated to the at least one iterative workload by evaluating, for each iteration of the at least one iterative workload, a representation of a relationship between: (i) the current allocation of the at least one resource allocated to the at least one iterative workload, (ii) a performance metric, wherein a value of the performance metric is changed for at least one iteration of the at least one iterative workload and is maintained at least for the at least one iteration, and (iii) the current performance of the at least one iterative workload, wherein the determined adjustment to the current allocation of the at least one resource allocated to the at least one iterative workload is controlled to be between a first upper limit value and a second independent lower limit value and wherein one or more of the first upper limit value and the second independent lower limit value is: (i) changed for a plurality of iterations of the at least one iterative workload and (ii) determined using a percentage of the current allocation of the at least one resource; and
initiating an application of the determined adjustment to the current allocation of the at least one resource for the at least one iterative workload.

9. The apparatus of claim 8 , wherein the performance metric comprises a nominal value of a predefined service metric.

10. The apparatus of claim 8 , wherein the current performance of the at least one iterative workload comprises a current value of a variable that tracks a given predefined service metric of the at least one iterative workload.

11. The apparatus of claim 10 , wherein a current error of the variable that tracks the given predefined service metric of the at least one iterative workload comprises a difference between the current value of the given predefined service metric and a corresponding predefined target value for the given predefined service metric.

12. The apparatus of claim 8 , wherein the at least one iterative workload comprises one workload and wherein the representation comprises an analytic representation of a quadratic representation of the relationship.

13. The apparatus of claim 8 , wherein the at least one iterative workload comprises a plurality of workloads and wherein the representation comprises a quadratic representation of the relationship.

14. The apparatus of claim 8 , wherein a sum of the at least one resource is constrained to an available amount of the at least one resource.

15. A non-transitory processor-readable storage medium having stored therein program code of one or more software programs, wherein the program code when executed by at least one processing device causes the at least one processing device to perform the following steps:
obtaining a current performance of at least one iterative workload;
determining an adjustment to a current allocation of at least one resource allocated to the at least one iterative workload by evaluating, for each iteration of the at least one iterative workload, a representation of a relationship between: (i) the current allocation of the at least one resource allocated to the at least one iterative workload, (ii) a performance metric, wherein a value of the performance metric is changed for at least one iteration of the at least one iterative workload and is maintained at least for the at least one iteration, and (iii) the current performance of the at least one iterative workload, wherein the determined adjustment to the current allocation of the at least one resource allocated to the at least one iterative workload is controlled to be between a first upper limit value and a second independent lower limit value and wherein one or more of the first upper limit value and the second independent lower limit value is: (i) changed for a plurality of iterations of the at least one iterative workload and (ii) determined using a percentage of the current allocation of the at least one resource; and
initiating an application of the determined adjustment to the current allocation of the at least one resource for the at least one iterative workload.

16. The non-transitory processor-readable storage medium of claim 15 , wherein the performance metric comprises a nominal value of a predefined service metric.

17. The non-transitory processor-readable storage medium of claim 15 , wherein the current performance of the at least one iterative workload comprises a current value of a variable that tracks a given predefined service metric of the at least one iterative workload.

18. The non-trans

Assignees

• Emc Ip Holding Co Llc

Inventors

• Calmon Tiago Salviano
• Sousa Eduardo Vera
• Gottin Vinícius Michel
• Bhaya Amit
• Diene Oumar
• Passoni Jonathan Ferreira

Classifications

• G06F9/50 Primary

  Allocation of resources, e.g. of the central processing unit [CPU] · CPC title

• G06F9/445

  Program loading or initiating (bootstrapping G06F9/4401; security arrangements for program loading or initiating G06F21/57) · CPC title

• G06F9/45541

  Bare-metal, i.e. hypervisor runs directly on hardware · CPC title

• G06F2209/501

  Performance criteria · CPC title

• G06N20/00

  Machine learning · CPC title