IP Library Granted Patent US 11,288,102
Granted Patent B2
US 11,288,102 · App. 15/689,148 · Granted Mar 29, 2022

Modifying resources for composed systems based on resource models

Inventors: Jeffrey R. Hamilton (Pittsboro, NC); James J. Parsonese (Cary, NC); Pravin Patel (Cary, NC)
Assignee: LENOVO ENTERPRISE SOLUTIONS (SINGAPORE) PTE. LTD.
G06F9/505G06F9/5083H04L67/1008H04L67/1031G06F15/173H04L67/1002
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 11,288,102
App. No.
15/689,148
Granted
Mar 29, 2022
Kind
B2
Abstract

Modifying resources for composed systems based on resource models including receiving a workload for execution on a composed system; extracting workload characteristics from the workload; matching the workload characteristics to a resource model, wherein the resource model comprises an initial configuration of compute elements for the composed system and a configuration modification to the initial configuration of the compute elements as the workload executes; composing the composed system using the initial configuration of compute elements described by the resource model, wherein the composed system comprises a subset of compute elements from a resource pool of compute elements; and executing, based on the resource model, the workload using the composed system, including modifying the initial configuration of the compute elements according to the resource model.

Claims (48)

1. A method comprising:

by program instructions on a computing device,

receiving a workload for execution on a composed system;

extracting workload characteristics from the workload;

matching the workload characteristics to one of a plurality of resource models stored in a resource model repository, wherein the resource model comprises an initial configuration of compute elements for the composed system and a configuration modification to the initial configuration of the compute elements as the workload executes, wherein the resource model further comprises a trigger point indicating when, during the execution of the workload, the initial configuration of compute elements is to be modified;

composing the composed system from a subset of compute elements in a resource pool of compute elements, wherein the subset of compute elements satisfies requirements of the initial configuration of compute elements described by the resource model; and

executing, based on the resource model, the workload using the composed system, including modifying, during the execution of the workload and in response to detecting the trigger point, the initial configuration of the compute elements according to the configuration modification of the resource model.

2. The method of claim 1 , further comprising:

generating a tuned resource model based on a comparison of the resource model to a monitored workload performance of the workload executing on the composed system.

3. The method of claim 1 , wherein receiving the workload for execution on the composed system comprises receiving a workload application and workload data,

wherein extracting workload characteristics of the workload comprises extracting data characteristics from the workload data and extracting application characteristics from the workload application, and

wherein matching the characteristics to a resource model comprises matching the application characteristics and the data characteristics to the resource model.

4. The method of claim 1 , wherein matching the workload characteristics to the resource model comprises selecting the resource model based on comparing the workload characteristics to a plurality of resource model workload characteristics.

5. The method of claim 1 , wherein modifying the initial configuration of the compute elements according to the resource model comprises one selected from a group consisting of:

removing a communications link on an interconnect fabric between at least two of the subset of compute elements, and

adding a communications link on an interconnect fabric between an additional compute element and the subset of compute elements.

6. The method of claim 1 , wherein the subset of compute elements comprises a hardware compute element and a software compute element.

7. An apparatus comprising a computing device, a computer processor, and a computer memory operatively coupled to the computer processor, the computer memory having disposed within it computer program instructions that, when executed by the computer processor, cause the apparatus to carry out the steps of:

receiving a workload for execution on a composed system;

extracting workload characteristics from the workload;

matching the workload characteristics to one of a plurality of resource models stored in a resource model repository, wherein the resource model comprises an initial configuration of compute elements for the composed system and a configuration modification to the initial configuration of the compute elements as the workload executes, wherein the resource model further comprises a trigger point indicating when, during the execution of the workload, the initial configuration of compute elements is to be modified;

composing the composed system from a subset of compute elements in a resource pool of compute elements, wherein the subset of compute elements satisfies requirements of the initial configuration of compute elements described by the resource model; and

executing, based on the resource model, the workload using the composed system, including modifying, during the execution of the workload and in response to detecting the trigger point, the initial configuration of the compute elements according to the configuration modification of the resource model.

8. The apparatus of claim 7 , wherein the computer program instructions further cause the apparatus to carry out the step of:

generating a tuned resource model based on a comparison of the resource model to a monitored workload performance of the workload executing on the composed system.

9. The apparatus of claim 7 , wherein receiving the workload for execution on the composed system comprises receiving a workload application and workload data,

wherein extracting workload characteristics of the workload comprises extracting data characteristics from the workload data and extracting application characteristics from the workload application, and

wherein matching the characteristics to a resource model comprises matching the application characteristics and the data characteristics to the resource model.

10. The apparatus of claim 7 , wherein matching the workload characteristics to the resource model comprises selecting the resource model based on comparing the workload characteristics to a plurality of resource model workload characteristics.

11. The apparatus of claim 7 , wherein modifying the initial configuration of the compute elements according to the resource model comprises one selected from a group consisting of:

removing a communications link on an interconnect fabric between at least two of the subset of compute elements, and

adding a communications link on an interconnect fabric between an additional compute element and the subset of compute elements.

12. The apparatus of claim 7 , wherein the subset of compute elements comprises a hardware compute element and a software compute element.

13. A computer program product including a non-volatile computer readable storage medium, the computer program product comprising computer program instructions that, when executed, cause a computer to carry out the steps of:

receiving a workload for execution on a composed system;

extracting workload characteristics from the workload;

matching the workload characteristics to one of a plurality of resource models stored in a resource model repository, wherein the resource model comprises an initial configuration of compute elements for the composed system and a configuration modification to the initial configuration of the compute elements as the workload executes, wherein the resource model further comprises a trigger point indicating when, during the execution of the workload, the initial configuration of compute elements is to be modified;

composing the composed system from a subset of compute elements in a resource pool of compute elements, wherein the subset of compute elements satisfies requirements of the initial configuration of compute elements described by the resource model; and

executing, based on the resource model, the workload using the composed system, including modifying, during the execution of the workload and in response to detecting the trigger point, the initial configuration of the compute elements according to the configuration modification of the resource model.

14. The computer program product of claim 13 , wherein the computer program instructions further cause the computer to carry out the step of:

generating a tuned resource model based on a comparison of the resource model to a monitored workload performance of the workload executing on the composed system.

15. The computer program product of claim 13 , wherein receiving the workload for execution on the composed system comprises receiving a workload application and workload data,

wherein extracting workload characteristics of the workload comprises extracting data characteristics from the workload data and extracting application characteristics from the workload application, and

wherein matching the characteristics to a resource model comprises matching the application characteristics and the data characteristics to the resource model.

16. The computer program product of claim 13 , wherein matching the workload characteristics to the resource model comprises selecting the resource model based on comparing the workload characteristics to a plurality of resource model workload characteristics.

17. The computer program product of claim 13 , wherein modifying the initial configuration of the compute elements according to the resource model comprises one selected from a group consisting of:

removing a communications link on an interconnect fabric between at least two of the subset of compute elements, and

adding a communications link on an interconnect fabric between an additional compute element and the subset of compute elements.

Assignments (3)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jan 10, 2025
From: LENOVO GLOBAL TECHNOLOGIES INTERNATIONAL LIMITED
To: LENOVO GLOBAL TECHNOLOGIES SWITZERLAND INTERNATIONAL GMBH
Reel/Frame 069869/0614 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded May 8, 2023
From: LENOVO ENTERPRISE SOLUTIONS (SINGAPORE) PTE.LTD.
To: LENOVO GLOBAL TECHNOLOGIES INTERNATIONAL LTD.
Reel/Frame 063564/0925 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Aug 29, 2017
From: HAMILTON, JEFFREY R.; PARSONESE, JAMES J.; PATEL, PRAVIN
To: LENOVO ENTERPRISE SOLUTIONS (SINGAPORE) PTE. LTD.
Reel/Frame 043708/0978 →
Continuity (1)
Related Publication 20190065256A1 · Feb 28, 2019
Cited By (2)
US 12,204,946 US 12,423,141