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

System and method for decentralized scheduling of autonomous flow engine load balancing clusters

A process engine and load balancing technology, applied in the field of cluster computing in the field of information technology, can solve problems such as single-port performance bottleneck, long response time, single-point failure, etc., to save investment costs, fast request response, and good customer experience.

Active Publication Date: 2015-03-04
京北方信息技术股份有限公司
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] The server side of the present invention does not set up a general scheduler, but adopts a multi-layer autonomous structure and a decentralized scheduling mode. Specifically, the server side includes several cluster node groups, each cluster node group has complete functions and autonomy, and is distributed and scheduled by the client, which can Effectively solve the problems of long response time, single-port performance bottleneck, and single-point failure of traditional load balancers

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
  • System and method for decentralized scheduling of autonomous flow engine load balancing clusters
  • System and method for decentralized scheduling of autonomous flow engine load balancing clusters
  • System and method for decentralized scheduling of autonomous flow engine load balancing clusters

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0056] The decentralized scheduling autonomous load balancing cluster system structure of the present invention is as follows: figure 1 As shown, the system is divided into N cluster node groups, each cluster node group is arranged with M cluster nodes, each node corresponds to a server computer, K engines are started on each cluster node, and each engine corresponds to an independent thread , the number of engines on nodes in different node groups can be different, but the number of engines on different nodes in the same node group must be the same, and the serial numbers of engines between node groups must be continuous. figure 1 The dashed arrows in indicate a synchronous call. The system is provided with a rotator at the client side.

[0057] N cluster node groups represent N engine line program number segments, and each cluster node group is responsible for calculating the cluster task request of the segment it belongs to. For example, the first group is responsible for...

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 system and a method for decentralized scheduling of autonomous flow engine load balancing clusters and belongs to the technology of cluster computing in the technical field of information. In the system and the method, multilayer clustering is performed on a server flow engine, the server flow engine is divided into a plurality of cluster node groups, each cluster node group is divided into a plurality of cluster nodes, and each cluster node is subdivided into a plurality of flow engines; and different cluster node groups carry out autonomy, a total scheduler is cancelled, and the cluster nodes in the same cluster node group are mutually spared by adopting a heartbeat detection method. The conventional total scheduling function is transferred from a server system and is scattered into a client system, and the client system calculates a requested service provider by adopting a rotation method according to a flow engine list so as to reduce the scheduling pressure of a server. According to the system and the method, the problems of single-port bottlenecks and single-point failure of the conventional load balancing clusters are effectively solved through the decentralized scheduling, and the response speed and processing capacity of the clusters are improved.

Description

technical field [0001] The invention relates to a process engine load balancing cluster system and method, and belongs to the cluster computing field in the field of information technology. Background technique [0002] High-performance cluster is the key technology of large-scale concurrent information system, which directly affects the availability of large-scale information system. In particular, the process engine of the workflow management system needs a cluster system with fast response and strong scalability. A traditional load balancing cluster system generally has a general scheduler, and the general scheduler is responsible for allocating requests from external systems to each processing unit. [0003] This clustering approach has the following deficiencies: [0004] 1) The client call is first passed to the general scheduler of the cluster system, and then the general scheduler of the cluster system forwards the request to the corresponding cluster node accordin...

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): H04L12/70H04L29/08H04L12/803
Inventor 孙刚高昊江
Owner 京北方信息技术股份有限公司
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