Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

A multi-agent cluster computing method

A multi-agent and computing method technology, applied in the field of cluster computing, can solve problems such as cluster redeployment, user maintenance burden, and interruption of computing processes

Active Publication Date: 2020-07-07
RES CENT FOR ECO ENVIRONMENTAL SCI THE CHINESE ACAD OF SCI
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the existing cluster computing method has the following disadvantages: (1) when a hardware failure occurs in a certain machine in the cluster, the entire cluster needs to be restarted, thereby interrupting the entire computing process; (2) when a machine is dynamically added or deleted in the cluster, The entire cluster needs to be redeployed and restarted
Both of them bring a huge burden to user maintenance, and it is necessary to find a new method suitable for multi-agent cluster computing

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
  • A multi-agent cluster computing method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0055] 1) Cluster fault-tolerant processing

[0056] Hadoop has a good machine failure handling (fault tolerance) mechanism, that is, it can monitor and adjust in time afterwards. Google believes that machine failures are normal, so fault handling is fully considered in the MapReduce system. This mainly includes two aspects: one is to have the ability to monitor the task status, and to be able to obtain the status of each task execution in time; the other is to base on the type of task (Map task or Reduce task) and the status of subtask execution (executing , Completed, Failed), determine the follow-up processing strategy. For example, if a TaskTracker fails, the JobTracker will transfer the tasks it is responsible for to another idle TaskTrack to run again.

[0057] 2) NameNode failure

[0058] When the data node does not receive the heartbeat message of the NameNode node for more than two default intervals (1200s), the NameNode is considered to be faulty. Since the data s...

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 invention relates to a multi-agent cluster operation method comprising the steps of building of a cluster environment: using 3 commercial PC machined, wherein one machine is used as a master of acluster, namely a NameNode, the other machine is used as a Secondary NameNode, serving as the backup of the master NameNode, the master NameNode is responsible for managing a whole file system, maintaining and updating block and storage information of files, and monitoring a health state of the file system, and the residual cluster is used as a TaskTracker and a DataNode at the same time; classification of the multi-agent; building of the multi-agent: comprising recognition of microcosmic individuals in the system, and construction of an internal model of the agent; building of a multi-agent system which comprises four parts, namely environment, structure, communication and coordination parts; and mapping an agent interaction process of an MAS to a specific task of an MR, namely expressingas that MAS.iter.agent is equal to MR.job.task, wherein MAS is the multi-agent system, and MapReduce is abbreviated as MR; and visualization of the multi-agent.

Description

technical field [0001] The invention relates to a cluster computing method, in particular to a multi-agent cluster computing method. Background technique [0002] Multi-Agents System (MAS, Multi-Agents System) is a thinking method and tool for analyzing and simulating complex systems. Based on the above ideas, build a model for a complex system. The micro-individuals in the model can have rich attributes and close to real decision-making logic and behavioral characteristics. Through the interaction between individuals and the interaction between individuals and the environment, the macroscopic characteristics of the system emerge. , so as to establish a bridge between the micro and macro. Multi-agent simulation modeling is flexible, natural, and individual attributes and behaviors are not limited. It is especially suitable for the study of complex systems composed of micro-individuals with certain intelligence, and has been valued by biology, economics, sociology and other ...

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): H04L29/08
CPCH04L67/10
Inventor 马金锋马梅徐强饶凯锋
Owner RES CENT FOR ECO ENVIRONMENTAL SCI THE CHINESE ACAD OF SCI
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com