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

Anomaly detection method and system for computing-intensive parallel tasks

An anomaly detection, intensive technology, applied in the direction of concurrent instruction execution, machine execution device, generation of response errors, etc., can solve problems such as not suitable for computationally intensive tasks

Active Publication Date: 2016-09-21
TSINGHUA UNIV
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This type of computing framework is reasonable for the detection of abnormal nodes of data-intensive tasks (such as Word Count), but it is not suitable for computing-intensive 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
  • Anomaly detection method and system for computing-intensive parallel tasks
  • Anomaly detection method and system for computing-intensive parallel tasks
  • Anomaly detection method and system for computing-intensive parallel tasks

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0025] Embodiments of the present invention are described in detail below, examples of which are shown in the drawings, wherein the same or similar reference numerals designate the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the figures are exemplary only for explaining the present invention and should not be construed as limiting the present invention. On the contrary, the embodiments of the present invention include all changes, modifications and equivalents coming within the spirit and scope of the appended claims.

[0026] In the description of the present invention, it should be understood that the terms "first", "second" and so on are used for descriptive purposes only, and cannot be interpreted as indicating or implying relative importance. In the description of the present invention, it should be noted that unless otherwise specified and limited, the terms "connected" and "connect...

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 present invention proposes an anomaly detection method for computing-intensive parallel tasks, including the following steps: analyzing the program machine code to obtain the function coverage information of the program machine code; selecting a task progress tracking function according to the function coverage information, and tracking the task progress The entry of the function is instrumented to obtain the instrumented program machine code; multiple computing nodes execute multiple tasks of the job through the instrumented program machine code; obtain multiple tracking information of multiple computing nodes; Perform cluster analysis on multiple tracking information to obtain cluster analysis results; judge abnormal nodes in multiple computing nodes according to cluster analysis results; judge whether all the executed tasks are completed, and if they are all completed, end the abnormal node. detection. The method of the invention can efficiently and accurately detect abnormal nodes. The invention also proposes an anomaly detection system for computing-intensive parallel tasks.

Description

technical field [0001] The invention relates to the field of distributed and parallel computing, in particular to a method and system for abnormality detection of computing-intensive parallel tasks. Background technique [0002] The development of cloud computing provides users with abundant computing resources. A data center usually has thousands of servers to process massive parallel tasks to support user applications (such as information retrieval, statistical analysis, and scientific computing, etc.). At present, the scale of traditional server clusters has also been further expanded, generally having thousands of CPU cores. Under such a large-scale architecture, effective and stable computing models and frameworks are particularly critical. [0003] In the SPMD (Single Program Multiple Data) parallel processing computing framework, users use computing resources by submitting jobs. Each job is divided into multiple tasks with the same computing scale and executed on ea...

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 Patents(China)
IPC IPC(8): G06F11/07G06F9/38
Inventor 武永卫郑纬民陈康郭维超
Owner TSINGHUA UNIV