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

Method for improving performance tuning speed of distributed system

A distributed system and performance technology, applied in transmission systems, biological neural network models, multi-programming devices, etc., can solve problems such as long measurement time, limited search capability, and reduced measurement time system performance.

Active Publication Date: 2009-12-23
CERTUS NETWORK TECHNANJING
View PDF0 Cites 20 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The above methods have their own advantages and disadvantages, and can improve the overall performance of the system to a certain extent. The common disadvantage is that the measurement time is too long
The experimental results given did reduce the measurement time and improve the system performance, but in the examples given, the search space is very suitable for discrete situations, but after being extended to continuous spaces, the problem will emerge
Because its search space is valid for a specific parameter configuration, but its search capability is limited

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
  • Method for improving performance tuning speed of distributed system
  • Method for improving performance tuning speed of distributed system
  • Method for improving performance tuning speed of distributed system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0021] Premise: Assume that the distributed computing system we are tuning now contains n parameters {p 1 ,p 2 ,...,p n}, parameter p i The value range of is: [ p i ,p i ], the algorithm is as follows:

[0022] (1) Put all the [ p i ,p i ] linearly quantized into [0, 100].

[0023] (2) Randomly and uniformly take 200 sets of parameters {v in [0, 100] 1，1 , v 2，1 ,...,v n，1}{v 12 , v 22 ,...,v n2},...,{v 1，200 , v 2，200 ,...,v n，200} and linearly quantize it back to the original interval [ p i ,p i ].

[0024] (3) Test the output under the above 200 sets of parameters, and set the response time and throughput rate as (t 1，1 , r 2，1 ), (t 1，2 , r 2，2 ),…, (t 1，200 , r 2，200 ) Use the 200 sets of output results of the test to train a three-layer neural network model with two hidden layers (see Appendix 1 for the principle and analysis process of the model, and also see Neural Network, Chapter 4, page 111, Neural Network edited by Simon Haykin how to us...

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

A method for improving performance tuning speed of a distributed system belongs to the technical category of performance optimization of a distributed computer system. The method is characterized by comprising the steps of quantizing the parameters with respective numeric area as a uniform parameter set in the distributed system including a web page server, an application programme server and a database server and composed of three-layer servers, obtaining the parameter set through uniform sampling, training by using a three-layer neural network model which comprises two hidden layers and of which each node is a logistic function, then repeatedly sampling, testing new parameters in the neural network to obtain rough response time and rough throughput rate, determining the sequencing performance curve and noise grade of the neural network based on the requirements of the system performance optimization, and finally obtaining the parameters of a regression function according to the set relative performance requirements so as to obtain the desired optimal parametric solution. The invention has the advantages that the invention improves the system performance and synchronously reduces test time.

Description

technical field [0001] The invention belongs to the field of computer performance optimization, in particular to the field of performance optimization of a three-layer server (web server, application program server and database server) system. Background technique [0002] With the continuous increase in the number of online users and the continuous improvement of users' application requirements, many challenges have been brought to the field of distributed system performance tuning. Cluster computing, grid computing, and Saibai platform all use complex distributed systems to meet the different needs of individuals, businesses, and governments in various fields and levels. Today, the cloud computing technology, which is widely respected by the industry and academia and is dedicated to completely changing the existing computer operation mode, has emerged as the times require. Its core idea is to put data in a huge server "cloud" background, and the end users at the front des...

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): G06F9/46G06N3/02H04L29/08
Inventor 曹军威张帆
Owner CERTUS NETWORK TECHNANJING
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