hadoop-hdfs-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Ming Ma (JIRA)" <j...@apache.org>
Subject [jira] [Created] (HDFS-9259) Make SO_SNDBUF size configurable at DFSClient side for hdfs write scenario
Date Fri, 16 Oct 2015 15:50:05 GMT
Ming Ma created HDFS-9259:
-----------------------------

             Summary: Make SO_SNDBUF size configurable at DFSClient side for hdfs write scenario
                 Key: HDFS-9259
                 URL: https://issues.apache.org/jira/browse/HDFS-9259
             Project: Hadoop HDFS
          Issue Type: Improvement
            Reporter: Ming Ma


We recently found that cross-DC hdfs write could be really slow. Further investigation identified
that is due to SendBufferSize and ReceiveBufferSize used for hdfs write. The test is to do
"hadoop -fs -copyFromLocal" of a 256MB file across DC with different SendBufferSize and ReceiveBufferSize
values. The results showed that c much faster than b; b is faster than a.

a. SendBufferSize=128k, ReceiveBufferSize=128k (hdfs default setting).
b. SendBufferSize=128K, ReceiveBufferSize=not set(TCP auto tuning).
c. SendBufferSize=not set, ReceiveBufferSize=not set(TCP auto tuning for both)

HDFS-8829 has enabled scenario b. We would like to enable scenario c to make SendBufferSize
configurable at DFSClient side. Cc: [~cmccabe] [~He Tianyi] [~kanaka] [~vinayrpet].



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message