You are viewing a plain text version of this content. The canonical link for it is here.

Posted to user@atlas.apache.org by "lemonzone2010@163.com" <le...@163.com> on 2020/03/28 05:10:29 UTC

How to improve performance when Atlas is configuring with HBase cluster

Hi
    My current Atlas:
    1. version: 1.1.0
    2. Elasticsearch  + HBase( local, standalone mode, deployed on Atlas server)

For Configuring Atlas server to use the High Availability feature, i change local HBase to HBase cluster in HA mode with HDFS. I got this performance metrics by enabling performance logs atlas-log4j.xml.
How to improve performance ? 
    
operation HBase in standalone mode
avg cost(ms) HBase cluster in HA mode
avg cost(ms) avg cost rate(%) 
ENTITY_DELETE_V2 31783.1 52625.7 -66 
ENTITY_FULL_UPDATE_V2 3371.84 14730.9 -337 
ENTITY_CREATE_V2 11835.6 28876.4 -144 
createOrUpdate 15010.5 27610.9 -84 




lemonzone2010@163.com