Unlock instant, AI-driven research and patent intelligence for your innovation.

Task scheduling method, system and equipment

A task scheduling and task technology, applied in the computer field, can solve problems such as affecting the processing speed of computing tasks, and achieve the effect of improving the processing speed

Pending Publication Date: 2021-07-16
ALIBABA GRP HLDG LTD
View PDF0 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This scheduling method is simple, and it is difficult to find suitable computing nodes for computing tasks, which affects the processing speed of computing tasks

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Task scheduling method, system and equipment
  • Task scheduling method, system and equipment
  • Task scheduling method, system and equipment

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0085] Through the observation of a large number of artificial intelligence model training and high-performance computing, the inventor found that although the GPU has powerful parallel processing capabilities, it can greatly speed up the data processing process, but the computing power required for real large-scale operations is much higher than that of a single machine. The GPU operation of a single card is not enough, which requires a single machine multi-card or even multi-machine multi-card cooperative training. There are many factors that affect the processing speed in the case of single-machine multi-card or multi-machine multi-card cooperative processing, mainly in the following aspects:

[0086] 1. The communication bandwidth between different GPUs on the same host is different, and different choices will bring different application performance. On the same host, communication occurs between GPUs, and due to the physical topology of GPUs, the I / O communication overhea...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The embodiment of the invention provides a task scheduling method, system and equipment. The method comprises the following steps: determining a heterogeneous resource combination from each computing node in a heterogeneous computing cluster according to a to-be-scheduled task; determining a target heterogeneous resource combination from the determined heterogeneous resource combinations according to the runtime information of each heterogeneous resource combination; and scheduling the to-be-scheduled task to a target computing node where the target heterogeneous resource combination is located, so that the target computing node allocates the to-be-scheduled task to the target heterogeneous resource combination for processing. According to the scheme provided by the embodiment of the invention, the appropriate target heterogeneous resource combination can be determined according to the actual operation condition of each heterogeneous resource combination, so that the processing speed of the to-be-scheduled task is improved.

Description

technical field [0001] The present application relates to the field of computer technology, in particular to a task scheduling method, system and equipment. Background technique [0002] In the fields of high-performance computing and artificial intelligence, it is usually necessary to accelerate computing tasks based on heterogeneous computing clusters. Usually, a heterogeneous computing cluster includes multiple computing nodes, and each computing node is composed of at least one central processing unit (CPU) and multiple heterogeneous computing chips. Wherein, the heterogeneous computing chip may be GPU (Graphics Processing Unit, image processor), FPGA (Field Programmable Gate Array, field programmable gate array). [0003] Currently, the scheduling strategy for heterogeneous computing clusters is to schedule computing tasks to available computing nodes for processing. This scheduling method is simple, and it is difficult to find a suitable computing node for the comput...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06F9/48G06F9/50
CPCG06F9/4881G06F9/5027
Inventor 车漾张凯徐晓舟李青坪
Owner ALIBABA GRP HLDG LTD