Information handling systems and methods to provide workload remediation based on workload performance metrics and contextual information

Patent metadata

Field	Value
Publication number	US-2023125489-A1
Application number	US-202117510575-A
Country	US
Kind code	A1
Filing date	Oct 26, 2021
Priority date	Oct 26, 2021
Publication date	Apr 27, 2023
Grant date	—

Abstract

The present disclosure provides various embodiments of information handling systems and related methods to provide workload remediation on client devices running multiple concurrent workloads. More specifically, the present disclosure provides software services and computer-implemented methods that utilize workload performance metrics and contextual information to provide workload remediation for each workload/application included within a user's workspace. The disclosed embodiments provide an automated iterative remediation framework, which identifies degradation of workload performance metrics of each workload, takes one or more corrective actions to remediate the performance degradation based on a set of observed states obtained for each workload, measures the efficacy of each corrective action using a weighted scoring function, and improves the workload performance for each workload by selecting the corrective action that optimizes the weighted scoring function for the set of observed states.

First claim

What is claimed is:

1 . An information handling system (IHS), comprising:
a computer readable storage device that stores workload remediation services, which are executable to improve workload performance of a plurality of workloads included within a user's workspace; and a host processor, wherein for each workload within the plurality of workloads, the host processor executes the workload remediation services to:
identify degradation of at least one workload performance metric corresponding to the workload;
take one or more corrective actions to remediate the degradation of the at least one workload performance metric based on a set of observed states obtained for the workload;
measure the efficacy of each corrective action using a weighted scoring function; and
improve the workload performance for the workload by selecting a particular corrective action that optimizes the weighted scoring function for the set of observed states.

2 . The information handling system of claim 1 , wherein the plurality of workloads included within the user's workspace includes one or more of the following:
endpoint-native applications stored within the computer readable storage device and executed locally by the host processor; and cloud-native applications stored and executed remotely on a remote server or cloud instance.

3 . The information handling system of claim 2 , wherein one or more of the endpoint-native applications and the cloud-native applications is containerized.

4 . The information handling system of claim 1 , wherein the at least one workload performance metric is dependent on the workload.

5 . The information handling system of claim 4 , wherein the at least one workload performance metric comprises one or more of the following: frames per second (FPS), latency, bit rate, lag, throughput, and input/output operations per second (IOPS).

6 . The information handling system of claim 1 , wherein the weighted scoring function utilizes a variety of weights to account for a plurality of positive workload performance metrics and negative workload performance metrics associated with each corrective action.

7 . The information handling system of claim 6 , wherein the weighted scoring function calculates a difference between a summation of weighted positive workload performance metrics and a summation of weighted negative workload performance metrics.

8 . The information handling system of claim 7 , wherein the workload remediation services are further executed by the host processor to:
determine that the corrective action is worth taking, if the difference calculated by the weighted scoring function is a positive value; determine that the corrective action is not worth taking, if the difference calculated by the weighted scoring function is a negative value; and select the particular corrective action having the highest positive value.

9 . The information handling system of claim 1 , wherein the workload remediation services are further executed by the host processor to improve the workload performance for each workload by iterating through different observed states and corrective actions, measuring the efficacy of each corrective action via the weighted scoring function, and selecting the particular corrective action that optimizes the weighted scoring function for the set of observed states.

10 . The information handling system of claim 1 , wherein the plurality of workloads included within the user's workspace includes at least one containerized application.

11 . The information handling system of claim 10 , wherein the set of observed states include one or more of the following: telemetry data obtained from the containerized application, administrative configuration policies specified for the containerized application, contextual information about the IHS, and contextual information about the user.

12 . The information handling system of claim 10 , wherein the one or more corrective actions comprise one or more of the following: switching the workload to a different container type, switching a cloud resource tier used to process the workload, switching hardware resources allocated to a container running the workload, switching a security level set for the container running the workload, throttling power or performance for the container running the workload, boosting the power or performance for the container running the workload, throttling other containers or workloads currently running in the user's workspace, and hibernating other containers or workloads currently running in the user's workspace.

13 . A computer implemented method to provide workload remediation for a plurality of workloads included within a user's workspace, wherein the computer implemented method is performed by a host processor of an information handling system (IHS) executing program instructions stored within a computer readable storage device of the IHS, and wherein for each workload within the plurality of workloads, the computer implemented method comprises:
detecting one or more concurrently running workloads, wherein for each workload detected, the computer implemented method further comprises:
applying an action to the workload;
collecting data for the workload, the collected data comprising telemetry data obtained from the workload, administrative configuration policies specified for the workload and contextual information and the IHS and/or the user;
utilizing the collected data to obtain a set of observed states and workload performance metrics for the workload;
supplying the workload performance metrics to a weighted scoring function, which measures the efficacy of the action;
applying a new action to the workload and repeating said collecting, said utilizing and said supplying one or more times; and
improving workload performance for the workload by selecting a particular action that optimizes the weighted scoring function for the set of observed states.

14 . The computer implemented method of claim 13 , wherein the workload performance metrics include positive workload performance metrics and negative workload performance metrics, and wherein the weighted scoring function measures the efficacy of the action by:
applying weights to the positive workload performance metrics and the negative workload performance metrics; and determining a difference between the weighted positive workload performance metrics and the weighted negative workload performance metrics.

15 . The computer implemented method of claim 14 , wherein after said supplying and before said applying a new action, the computer implemented method further comprises:
determining that the action is: (i) worth taking if the difference is a positive value, and (ii) not worth taking if the difference is a negative value.

16 . The computer implemented method of claim 15 , wherein said improving workload performance for the workload comprises selecting the particular action that optimizes the weighted scoring function for the set of observed states by selecting the action that achieves the highest positive value.

17 . The computer implemented method of claim 13 , further comprising applying the selected action to the workload to improve at least one workload performance metric for the workload.

18 . The computer implemented method of claim 17 , wherein the at least one workload performance metric comprises one or more of the following: frames per second (FPS), latency, bit rate, lag, throughput, and input/output operations per second (IOPS).

19 . Th

Assignees

• Dell Products Lp

Inventors

• Vichare Nikhil M
• Iyer Vivek V

Classifications

• G06F11/3409

  for performance assessment · CPC title

• G06F9/5077 Primary

  Logical partitioning of resources; Management or configuration of virtualized resources (specific details on emulation or internal functioning of virtual machines G06F9/455) · CPC title

• G06F9/5083 Primary

  Techniques for rebalancing the load in a distributed system · CPC title

• G06F1/3203

  Power management, i.e. event-based initiation of a power-saving mode · CPC title

• G06F11/3433

  for load management (allocation of a server based on load conditions G06F9/505; load rebalancing G06F9/5083; redistributing the load in a network by a load balancer H04L67/1029) · CPC title