Data movement criteria for an information handling system

Patent metadata

Field	Value
Publication number	US-2025265130-A1
Application number	US-202418444932-A
Country	US
Kind code	A1
Filing date	Feb 19, 2024
Priority date	Feb 19, 2024
Publication date	Aug 21, 2025
Grant date	—

Abstract

An information handling system includes a memory and a processor. The memory stores a quality of service (QOS) table that includes parameters for a target application. The processor identifies the target application from multiple applications and receives a requested QoS for the target application. The processor determines a system workload class associated with the target application and determines a system level QoS associated with the target application. Based on the requested QoS, the system workload class, and the system level QoS, the processor determines a derived QoS for the target application. The requested QoS, the system workload class, the system level QoS, and the derived QoS are stored in the QoS table. Based on the derived QoS, the processor determines whether the target application is a candidate for migration from the information handling system to an offload compute device.

First claim

What is claimed is:

1 . An information handling system comprising:
a memory to store a quality of service (QOS) table, wherein the QoS table includes a plurality of parameters for a target application; a processor to communicate with the memory, the processor to:
identify the target application from a plurality of applications;
receive a requested QoS for the target application;
determine a system workload class associated with the target application;
determine a system level QoS associated with the target application;
based on the requested QoS, the system workload class, and the system level QoS, determine a derived QoS for the target application, wherein the requested QoS, the system workload class, the system level QoS, and the derived QoS are stored in the QoS table; and
based on the derived QoS, determine whether the target application is a candidate for migration from the information handling system to an offload compute device.

2 . The information handling system of claim 1 , wherein in response to the target application being a candidate to migration, the processor further to:
determine whether the offload compute device is able to meet the requested QoS for the target application; and in response to the offload compute device being able to meet the request QoS, provide the target application to the offload compute device.

3 . The information handling system of claim 2 , wherein the processor further to: receive data associated with target application from the offload compute device.

4 . The information handling system of claim 1 , wherein the processor further to: provide the target application to the offload compute device based on the migration to and from the offload compute device being able to meet a latency requirement for the target application.

5 . The information handling system of claim 1 , wherein the processor further to: assign power/latency quadrant for the target application, wherein the power/latency quadrant identifies a power usage and latency requirement for the target application based on a performance requirement of the target application.

6 . The information handling system of claim 1 , wherein the requested QoS for the target application includes a processor usage based on an algorithm complexity of the target application.

7 . The information handling system of claim 1 , wherein in response to the target application not being a candidate for migration, the processor to: assign the target application to a core of the processor for execution.

8 . The information handling system of claim 1 , wherein the system level QoS includes a maximum system power consumption, a maximum computation workload, a thread/application latency, and a schedule policy.

9 . A method comprising:
identifying, by a processor of an information handling system, a target application from a plurality of applications; receiving a requested quality of service (QOS) for the target application; determining a system workload class associated with the target application; determining a system level QoS associated with the target application; based on the requested QoS, the system workload class, and the system level QoS, determining a derived QoS for the target application; storing the requested QoS, the system workload class, the system level QoS, and the derived QoS in a QoS table of a memory of the information handling system; and based on the derived QoS, determining, by the processor, whether the target application is a candidate for migration from the information handling system to an offload compute device.

10 . The method of claim 9 , wherein in response to the target application being a candidate to migration, the method further comprising:
determining whether the offload compute device is able to meet the requested QoS for the target application; and in response to the offload compute device being able to meet the request QoS, providing the target application to the offload compute device.

11 . The method of claim 10 , further comprising: receiving data associated with target application from the offload compute device.

12 . The method of claim 10 further comprising: providing the target application to the offload compute device based on the migration to and from the offload compute device being able to meet a latency requirement for the target application.

13 . The method of claim 10 , further comprising: assigning power/latency quadrant for the target application, wherein the power/latency quadrant identifies a power usage and latency requirement for the target application based on a performance requirement of the target application.

14 . The method of claim 9 , wherein the requested QoS for the target application includes a processor usage based on an algorithm complexity of the target application.

15 . The method of claim 9 wherein in response to the target application not being a candidate for migration, the method further comprises: assigning the target application to a core of the processor for execution.

16 . The method of claim 9 , wherein the system level QoS includes a maximum system power consumption, a maximum computation workload, a thread/application latency, and a schedule policy.

17 . An information handling system comprising:
a memory to store a quality of service (QOS) table, wherein the QoS table includes a plurality of parameters for a target application; a processor to:
identify the target application from a plurality of applications;
receive a requested QoS for the target application;
determine a system workload class associated with the target application;
determine a system level QoS associated with the target application;
based on the requested QoS, the system workload class, and the system level QoS, determine a derived QoS for the target application, wherein the requested QoS, the system workload class, the system level QoS, and the derived QoS are stored in the QoS table;
based on the derived QoS, determine whether the target application is a candidate for migration from the information handling system to an offload compute device;
in response to the target application being a candidate to migration:
determine whether the offload compute device is able to meet the requested QoS for the target application; and
in response to the offload compute device being able to meet the request QoS, provide the target application to the offload compute device; and

in response to the target application not being a candidate for migration, assign the target application to a core of the processor for execution.

18 . The information handling system of claim 17 wherein the processor provides the target application to the offload compute device based on the migration to and from the offload compute device being able to meet a latency requirement for the target application.

19 . The information handling system of claim 17 , wherein the processor assigns power/latency quadrant for the target application, wherein the power/latency quadrant identifies a power usage and latency requirement for the target application based on a performance requirement of the target application.

20 . The information handling system of claim 17 , wherein the system level QoS includes a maximum system power consumption, a maximum computation workload, a thread/application latency, and a schedule policy.

Assignees

• Dell Products Lp

Inventors

• Khosrowpour Farzad
• Markow Mitchell
• Kwatra Ajay

Classifications

• G06F2209/509

  Offload · CPC title

• G06F9/5094

  where the allocation takes into account power or heat criteria (power management in computers in general G06F1/3203; thermal management in computers in general G06F1/206) · CPC title

• G06F2209/501

  Performance criteria · CPC title

• G06F9/5044

  considering hardware capabilities · CPC title

• G06F9/505

  considering the load · CPC title