Method, apparatus and computer-readable medium for application scheduling

Patent metadata

Field	Value
Publication number	US-2019121664-A1
Application number	US-201816162551-A
Country	US
Kind code	A1
Filing date	Oct 17, 2018
Priority date	Oct 20, 2017
Publication date	Apr 25, 2019
Grant date	—

Abstract

Embodiments of the present disclosure provide a method, apparatus and computer-readable medium for application scheduling. In accordance with embodiments of the present disclosure, influence factors related to scheduling of a plurality of applications between a plurality of processing units of a computing system are obtained, the plurality of applications being run by at least one of the plurality of processing units. Based on the obtained influence factors, a target application to be scheduled is selected from the plurality of applications and a first processing unit is selected from the plurality of processing units, the first processing unit being different from a second processing unit of the at least one processing unit running the target application. The target application is scheduled from the second processing unit to the first processing unit to continue running of the target application by the first processing unit.

First claim

I/we claim:

1 . A method of scheduling applications, comprising:
obtaining influence factors related to scheduling of a plurality of applications between a plurality of processing units of a computing system, the plurality of applications being run by at least one of the plurality of processing units; selecting, based on the obtained influence factors, a target application to be scheduled from the plurality of applications and a first processing unit from the plurality of processing units, the first processing unit being different from a second processing unit of the at least one processing unit running the target application; and scheduling the target application from the second processing unit to the first processing unit to continue running of the target application by the first processing unit.

2 . The method of claim 1 , wherein obtaining the influence factors comprises obtaining at least one of the following:
resource utilization of the plurality of processing units, resource utilization of each of the plurality of applications on each of the plurality of processing units running the application, and information related to a topology of the plurality of processing units, the information comprising at least one of a connection hop number and a type of connection between the plurality of processing units.

3 . The method of claim 1 , wherein selecting the target application and the first processing unit comprises:
selecting the target application and the first processing unit, such that differences between resource utilization of the plurality of processing units is to be reduced after the target application is scheduled to the first processing unit.

4 . The method of claim 1 , wherein selecting the target application and the first processing unit comprises:
selecting the target application and the first processing unit, such that differences between resource utilization of the plurality of processing units is to be increased after the target application is scheduled to the first processing unit.

5 . The method of claim 1 , wherein selecting the target application and the first processing unit comprises:
selecting the target application and the first processing unit, such that a predefined criteria for a cost of migrating data associated with the target application from the second processing unit to the first processing unit is to be satisfied, the cost being based on at least one of a connection hop number and a types of connections between the second processing unit and the first processing unit.

6 . The method of claim 1 , wherein selecting the target application and the first processing unit further comprises:
selecting the target application and the first processing unit further based on priorities of the plurality of applications, the priorities being at least related to service quality requirements of the plurality of applications.

7 . The method of claim 1 , wherein the plurality of processing units are comprised in a single machine of the computing system.

8 . The method of claim 1 , wherein the plurality of processing units are distributed over a plurality of machines of the computing system.

9 . The method of claim 1 , wherein the plurality of processing units is of a same type and selected from one of the following:
graphics processing units (GPUs), field programmable gate arrays (FPGAs), and central processing units (CPUs).

10 . An apparatus for scheduling applications, comprising:
a processor; and a memory coupled to the processor having instructions stored therein, the instructions, when executed by the processor, causing the apparatus to perform acts comprising: obtaining influence factors related to scheduling of a plurality of applications between a plurality of processing units of a computing system, the plurality of applications being run by at least one of the plurality of processing units; selecting, based on the obtained influence factors, a target application to be scheduled from the plurality of applications and a first processing unit from the plurality of processing units, the first processing unit being different from a second processing unit of the at least one processing unit running the target application; and scheduling the target application from the second processing unit to the first processing unit to continue running of the target application by the first processing unit.

11 . The apparatus of claim 10 , wherein obtaining the influence factors comprises obtaining at least one of the follows:
resource utilization of the plurality of processing units, resource utilization of each of the plurality of applications on each of the plurality of processing s running the application, and information related to a topology of the plurality of processing units, the information comprising at least one of a connection hop number and a types of connections among the plurality of processing units.

12 . The apparatus of claim 10 , wherein selecting the target application and the first processing unit comprises:
selecting the target application and the first processing unit, such that differences between resource utilization of the plurality of processing units is to be reduced after the target application is scheduled to the first processing unit.

13 . The apparatus of claim 10 , wherein selecting the target application and the first processing unit comprises:
selecting the target application and the first processing unit, such that differences between resource utilization of the plurality of processing units is to be increased after the target application is scheduled to the first processing unit.

14 . The apparatus of claim 10 , wherein selecting the target application and the first processing unit comprises:
selecting the target application and the first processing unit, such that a predefined criteria for a cost of migrating data associated with the target application from the second processing unit to the first processing unit is to be satisfied, the cost being based on at least one of a connection hop number and a type of connection between the second processing unit and the first processing unit.

15 . The apparatus of claim 10 , wherein selecting the target application and the first processing unit further comprises:
selecting the target application and the first processing unit further based on priorities of the plurality of applications, the priorities being at least related to service quality requirements of the plurality of applications.

16 . The apparatus of claim 10 , wherein the plurality of processing units are comprised in a single machine of the computing system.

17 . The apparatus of claim 10 , wherein the plurality of processing units are distributed over a plurality of machines of the computing system.

18 . The apparatus of claim 10 , wherein the plurality of processing units is of a same type and selected from one of the following:
graphics processing units (GPUs), field programmable gate arrays (FPGAs), and central processing units (CPUs).

19 . A computer-readable medium having computer readable instructions stored thereon, the computer-readable instructions, when executed, causing a device to perform:
obtaining influence factors related to scheduling of a plurality of applications between a plurality of processing units of a computing system, the plurality of applications being run by at least one of the plurality of processing units; selecting, based on the obtained influence factors, a target application to be

Assignees

• Emc Ip Holding Co Llc

Inventors

• Zhao Junping
• Guo Fan
• Wang Kun

Classifications

• G06F2209/482

  Application · CPC title

• G06F9/4881 Primary

  Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues · CPC title

• G06F9/4843 Primary

  by program, e.g. task dispatcher, supervisor, operating system · CPC title

• G06F9/505

  considering the load · CPC title

• Y02D10/00

  Energy efficient computing, e.g. low power processors, power management or thermal management · CPC title