Reconfigurable method of multi-cluster system

A cluster system, cluster technology, applied in the field of reconfigurable technology between multiple clusters

Inactive Publication Date: 2010-05-05
BEIHANG UNIV
View PDF0 Cites 36 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the prior art does not ef

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
  • Reconfigurable method of multi-cluster system
  • Reconfigurable method of multi-cluster system
  • Reconfigurable method of multi-cluster system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0039] The reconfigurable multi-cluster computing environment is composed of multiple member clusters 4, each cluster has an independent management domain and can work independently, the clusters are in an equal position, the combination and split of the multi-cluster computing environment does not need to change the management domain of each cluster , can be reorganized flexibly and conveniently, and has good scalability and usability. Its organizational structure is as image 3 shown.

[0040] If the multi-cluster computing environment consists of n member clusters C 1 , C 2 ,...,C n Composition, where C i Consisting of multiple homogeneous computing nodes 3, its recombinability can be described as: multi-cluster M={C i , the clusters in 1≤i≤n} can be arbitrarily combined into multiple cluster groups N 1 , N 2 ,…N m (1≤m≤n), and N j (1≤j≤m) is one of the non-empty subsets of M, if x∈N p , y∈N q , and p≠q, then there must be x≠y.

[0041] The management nodes 1 of...

Embodiment 2

[0101] The advantage of the exit processing of the Master cluster in Embodiment 1 is that it is easy to execute, but when the number of clusters is large or the cluster with a small number first finds that the Master cluster has exited, a large number of envelopes need to be transmitted on the network, and the system performance relatively low.

[0102] This embodiment improves the exit processing of the Master cluster in Embodiment 1, and obtains an optimized processing method. When a cluster P finds that the Master cluster no longer responds to requests, it initiates an election. refer to Figure 10 , the circles marked 0, 1, 2...7 in the figure represent multiple clusters, and the process is as follows:

[0103] 1) After the current Master cluster exits, it is first found that the cluster that the current Master cluster exits calls for elections, and the cluster that calls for elections sends an election message (election message) to all clusters with numbers larger than ...

Embodiment 3

[0112] For a distributed multi-cluster environment, if more than one or even all clusters discover that the Master fails at the same time and initiate elections at the same time, the network load will also increase. In view of this situation, this embodiment once again improves the methods for processing the exit of the Master cluster in Embodiment 1 and Embodiment 2.

[0113] refer to Figure 11 , the circles marked 0, 1, 2...7 in the figure represent multiple clusters, and the process is as follows:

[0114] 1) When a cluster called for election finds that the Master cluster no longer responds to the request, the cluster called for election sends an election message (election message) to all clusters with numbers larger than itself;

[0115] 2) If multiple clusters initiate elections at the same time, a cluster in the multi-cluster system will receive multiple election messages, and the cluster will only send multiple election messages to the cluster with a smaller number ...

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 discloses a reconfigurable method of a multi-cluster system and is used for solving the problem of flexible configuration such as combination of clusters, detachment of clusters. The multi-cluster system comprises a plurality of member clusters, wherein each member cluster comprises a management node and a calculation node; the management node is provided with a scheduler and a work manager; the management node is only interacted with one activated scheduler to schedule work; the member clusters are further divided into a master cluster and auxiliary clusters; and the master cluster is the member comprising the member clusters of the activated scheduler while the auxiliary cluster is the member cluster comprising no such scheduler. The method further comprises the following processing: processing one, according to system request, adding one cluster to be combined into the present multi-cluster; processing two, according to system request, determining a new master cluster when the master cluster quits from the present multi-cluster; and processing three, according to system request, removing an auxiliary cluster from the present multi-cluster, and rearranging the multi-cluster.

Description

technical field [0001] The invention relates to the technical field of high-performance computer clusters, in particular to the reconfigurable technology among multiple clusters. Background technique [0002] A cluster is a kind of computer system, which is connected by a group of loosely integrated computer software and hardware to complete computing work closely together. It has the advantages of low cost, easy maintenance, and flexible configuration. Generally, a cluster computer is better than a single computer. For example, the performance-price ratio of workstations or supercomputers is much higher. figure 1 It is the physical structure of a cluster, and multiple computers (management nodes 1 and computing nodes 3) are connected to form a network through a switch or other high-speed communication devices 2, that is, a simple computer cluster (hereinafter referred to as a cluster) is formed. figure 2 It is the logical structure of the cluster. [0003] However, with ...

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
IPC IPC(8): H04L29/06
Inventor 胡凯丁毅牛建伟陈陆佳那日苏张伟
Owner BEIHANG UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap