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

Energy consumption management method for SLA (service level agreement) based on heterogeneous MapReduce cluster

A technology for energy management and clustering, applied in the field of SLA-oriented energy management, can solve problems such as performance degradation and application response time unacceptable, and achieve the effect of reducing costs

Inactive Publication Date: 2015-08-19
HANGZHOU TIANKUAN TECH
View PDF0 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Fault tolerance means that lost data can be recovered quickly when the server goes down unexpectedly. Although shutting down nodes can reduce energy loss, it will lead to severe performance degradation in large-scale data parallel processing, making the response time of the application unacceptable.

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
  • Energy consumption management method for SLA (service level agreement) based on heterogeneous MapReduce cluster
  • Energy consumption management method for SLA (service level agreement) based on heterogeneous MapReduce cluster
  • Energy consumption management method for SLA (service level agreement) based on heterogeneous MapReduce cluster

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0026] Embodiment: a kind of SLA-oriented energy consumption management method based on heterogeneous MapReduce cluster, comprises the following steps:

[0027] (1) A hybrid data copy storage strategy is proposed for heterogeneous MapReduce clusters, which allows shutting down a large number of nodes while ensuring data integrity and cluster fault tolerance;

[0028] What needs to be explained in this embodiment is that the heterogeneous system is composed of multiple racks, the nodes in the racks are homogeneous, and the nodes between racks are heterogeneous, and there are m racks i(i =1,...,m) has N(i) nodes. Each node has different performance indicators and power consumption parameters, that is, running different applications will show different performance. Here, C(i,ω) represents the computing performance of the nodes in rack i when running the program ω, and E(i,ω) represents the energy consumption of the nodes in rack i when running the program ω per unit time. Here,...

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 an energy consumption management method for an SLA (service level agreement) based on a heterogeneous MapReduce cluster. First of all, a mixed data copy storage strategy is provided, and the mixed data copy storage strategy allows to close a great number of nodes and ensures the integrity of data and a fault tolerance of the cluster; then a response time prediction method based on historical record is designed, and the response time prediction method accurately estimates the response time of a program according to historical information of quantity and performance parameters and operating time of the server nodes with the relative error being mostly below 6%; and finally, energy consumption minimization is achieved through selectively closing part of the nodes, the real-time performance of the application program is ensured at the same time, and a heuristic node closing strategy is provided. Therefore, the real-time performance of a MapReduce application is ensured at a node closing strategy, and a lot of energy consumption is reduced at the same time.

Description

technical field [0001] The invention relates to the technical field of large-scale data processing, in particular to an SLA-oriented energy management method based on heterogeneous MapReduce clusters. Background technique [0002] With the continuous expansion of data center scale and the rapid expansion of network application services, the cost of energy consumption not only accounts for a large proportion of the total cost of data centers, but also limits the expansion and performance of infrastructure. According to statistics, data center Utilization is mostly maintained below 30%, but server energy consumption is high and most of it is wasted. In this context, researchers in industry and academia have proposed to reduce energy consumption by shutting down some servers during periods of low utilization. As the current mainstream large-scale data processing model, MapReduce has been widely used in major data centers such as Google, Yahoo! And Facebook, etc. The characte...

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): H04L12/24H04L29/08G06F17/30
CPCY02D10/00
Inventor 周天和蔡荣张元元卢晓飞
Owner HANGZHOU TIANKUAN TECH
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