An Efficient Dynamic Load-Balancing Large Scale Graph-Processing System
碩士 === 國立臺灣大學 === 資訊工程學研究所 === 106 === Since the introduction of pregel by Google, several large-scale graphprocessing systems have been introduced. These systems are based on the bulk synchronous parallel model or other similar models and use various strategies to optimize system performance. For e...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | en_US |
Published: |
2018
|
Online Access: | http://ndltd.ncl.edu.tw/handle/kpmh4n |
id |
ndltd-TW-106NTU05392118 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-106NTU053921182019-07-25T04:46:48Z http://ndltd.ncl.edu.tw/handle/kpmh4n An Efficient Dynamic Load-Balancing Large Scale Graph-Processing System 有效率的動態負載平衡圖形處理系統 Ming-Chia Kuo 郭明嘉 碩士 國立臺灣大學 資訊工程學研究所 106 Since the introduction of pregel by Google, several large-scale graphprocessing systems have been introduced. These systems are based on the bulk synchronous parallel model or other similar models and use various strategies to optimize system performance. For example, Mizan monitors the workload of each worker to determine whether the workload between the workers is balanced with respect to the execution time. If the workload is unbalanced among workers, Mizan migrates nodes from overloaded workers to under-loaded workers to balance the load among workers and minimize the total execution time. On the basis of Mizan’s migration plan, we implement a graph-processing system called GPSer with an efficient re-partitioning graph scheme. Our system uses statistical tools, e.g., coefficient of variation and correlation coefficient, to modify the migration plan and determine whether the workloads are balanced among all workers. Our system can accurately monitor current workloads and decide whether to migrate nodes among workers to balance the load. When imbalance arises, the workload of all workers can quickly converge to a balanced state, thereby enhancing the system performance. In experiment our system outperforms the state-of-the-art dynamic load-balancing graph processing-system, such as Mizan. Pangfeng Liu 劉邦鋒 2018 學位論文 ; thesis 29 en_US |
collection |
NDLTD |
language |
en_US |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 國立臺灣大學 === 資訊工程學研究所 === 106 === Since the introduction of pregel by Google, several large-scale graphprocessing systems have been introduced. These systems are based on the bulk synchronous parallel model or other similar models and use various strategies to optimize system performance. For example, Mizan monitors
the workload of each worker to determine whether the workload between the workers is balanced with respect to the execution time. If the workload is unbalanced among workers, Mizan migrates nodes from overloaded workers to
under-loaded workers to balance the load among workers and minimize the total execution time. On the basis of Mizan’s migration plan, we implement a graph-processing system called GPSer with an efficient re-partitioning graph scheme. Our system uses statistical tools, e.g., coefficient of variation and correlation coefficient, to modify the migration plan and determine whether the workloads are balanced among all workers. Our system can accurately monitor current workloads and decide whether to migrate nodes among workers to balance the load. When imbalance arises, the workload of all workers can quickly converge to a balanced state, thereby enhancing the system performance. In experiment our system outperforms the state-of-the-art dynamic load-balancing graph processing-system, such as Mizan.
|
author2 |
Pangfeng Liu |
author_facet |
Pangfeng Liu Ming-Chia Kuo 郭明嘉 |
author |
Ming-Chia Kuo 郭明嘉 |
spellingShingle |
Ming-Chia Kuo 郭明嘉 An Efficient Dynamic Load-Balancing Large Scale Graph-Processing System |
author_sort |
Ming-Chia Kuo |
title |
An Efficient Dynamic Load-Balancing Large Scale Graph-Processing System |
title_short |
An Efficient Dynamic Load-Balancing Large Scale Graph-Processing System |
title_full |
An Efficient Dynamic Load-Balancing Large Scale Graph-Processing System |
title_fullStr |
An Efficient Dynamic Load-Balancing Large Scale Graph-Processing System |
title_full_unstemmed |
An Efficient Dynamic Load-Balancing Large Scale Graph-Processing System |
title_sort |
efficient dynamic load-balancing large scale graph-processing system |
publishDate |
2018 |
url |
http://ndltd.ncl.edu.tw/handle/kpmh4n |
work_keys_str_mv |
AT mingchiakuo anefficientdynamicloadbalancinglargescalegraphprocessingsystem AT guōmíngjiā anefficientdynamicloadbalancinglargescalegraphprocessingsystem AT mingchiakuo yǒuxiàolǜdedòngtàifùzàipínghéngtúxíngchùlǐxìtǒng AT guōmíngjiā yǒuxiàolǜdedòngtàifùzàipínghéngtúxíngchùlǐxìtǒng AT mingchiakuo efficientdynamicloadbalancinglargescalegraphprocessingsystem AT guōmíngjiā efficientdynamicloadbalancinglargescalegraphprocessingsystem |
_version_ |
1719230001533943808 |