DocumentCode :
614035
Title :
Performance Analysis and Optimization of Map Only Left Outer Join
Author :
Ming Hao ; Wlodarczyk, T.W. ; Chunming Rong
Author_Institution :
Dept. of Electr. & Comput. Eng., Univ. of Stavanger, Stavanger, Norway
fYear :
2013
fDate :
25-28 March 2013
Firstpage :
625
Lastpage :
631
Abstract :
We studied the characteristics of HDFS, Distributed Cache and how algorithms of left outer join on map side had been implemented on the Hadoop platform. For the purpose of performance optimization we inspected several methods to control amount of map task. Further, according to the result of the experiment, we adjusted critical parameters. Based on these we significantly improved performance in comparison to other existing implementations and experiments.
Keywords :
cache storage; distributed databases; parallel processing; Hadoop distributed file system; Hadoop platform; distributed cache; left outer join algorithm; map side; map task amount; performance analysis; performance optimization; Algorithm design and analysis; Computer architecture; Conferences; Data processing; Dictionaries; Exponential distribution; Optimization; HDFS; Hadoop; left outer join; map only;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Advanced Information Networking and Applications Workshops (WAINA), 2013 27th International Conference on
Conference_Location :
Barcelona
Print_ISBN :
978-1-4673-6239-9
Electronic_ISBN :
978-0-7695-4952-1
Type :
conf
DOI :
10.1109/WAINA.2013.74
Filename :
6550466
Link To Document :
بازگشت