arrow
返回

Confluence: Speeding Up Iterative Distributed Operations by Key-Dependency-Aware Partitioning

delete2018-02-01
delete2
PRE
AI
梁锋 封面图
梁锋 (Feng Liang) *
F
Francis C. M. Lau
H
Heming Cui
DOI:10.1109/TPDS.2017.2756054delete
delete原文链接
delete原文求助
delete分享
delete收藏
摘要

摘要

En 中文
A typical shuffle operation randomly partitions data on many computers, generating possibly a significant amount of network traffic which often dominates a job's completion time. This traffic is particularly pronounced in iterative distributed operations where each iteration invokes a shuffle operation. We observe that data of different iterations are related according to the transformation logic of distributed operations. If data generated by the current iteration are partitioned to the computers where they will be processed in the next iteration, unnecessary shuffle network traffic between the two iterations can be prevented. We model general iterative distributed operations as the transform-and-shuffle primitive and define a powerful notion named Confluence key dependency to precisely capture the data relations in the primitive. We further find that by binding key partitions between different iterations based on the Confluence key dependency, the shuffle network traffic can always be reduced by a predictable percentage. We implemented the Confluence system. Confluence provides a simple interface for programmers to express the Confluence key dependency, based on which Confluence automatically generates efficient key partitioning schemes. Evaluation results on diverse real-life applications show that Confluence greatly reduces the shuffle network traffic, resulting in as much as 23 percent job completion time reduction.
Keyword:
Spark
shuffle
key dependency
iterative distributed operation
partitioning
AI总结

AI总结

对已上传原文的论文进行重点信息的提取,主要内容包括:简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。

期刊

IEEE Transactions on Parallel and Distributed Systems 封面图
IEEE Transactions on Parallel and Distributed Systems
IF:
6
论文数:
5.2K
被引数:
1.1W

机构

U
University of Hong Kong
学者数:
4.1W
论文数: 3.9W
被引数: 10.1W
引用论文

引用论文

err分享
err收藏
Traceme — indoor real-time location system
err2009-11-01
err0
PREAI
errPedro M. Mestre A. Silva; Maximino Paralta; Rafael Caldeirinha; Jorge Rodrigues; Carlos M.J.A. Serodio
err分享
err收藏
PDK2 induces cisplatin-resistance in lung adenocarcinoma via transcriptional regulation of CNNM3
err2018-12-04
err0
PREAI
errTinghua Hu; Shuo Yu; Yang Li; Hui Ren; Qian Ning; Jing Wang; Xuan Liang; Manxiang Li
err分享
err收藏
学者 查看更多内容