[ https://issues.apache.org/jira/browse/HBASE-26340?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17447414#comment-17447414 ]
Norbert Kalmár commented on HBASE-26340: ---------------------------------------- Made another patch, just tested it: {code:java} hbase:001:0> create 'emp', 'personal data', 'professional data' 2021-11-22T14:47:57,545 INFO [RPCClient-NioEventLoopGroup-1-2] client.AsyncHBaseAdmin: Operation: CREATE, Table Name: default:emp completed Created table emp Took 1.5477 seconds => Hbase::Table - emp hbase:002:0> list_regions 'emp' SERVER_NAME | REGION_NAME | START_KEY | END_KEY | SIZE | REQ | LOCALITY | -------------------------------- | ---------------------------------------------------- | ---------- | ---------- | ----- | ----- | ---------- | 192.168.1.75,16020,1637588865433 | emp,,1637588876835.d6b9eaddeedbf2342c8e834a1151cc2a. | | | 0.0MB | 0 | 0.0 | 1 rows Took 0.3317 seconds hbase:003:0> put 'emp','1','personal data:name','raju' Took 0.0650 seconds hbase:004:0> flush 'emp' Took 0.3037 seconds hbase:005:0> list_regions 'emp' SERVER_NAME | REGION_NAME | START_KEY | END_KEY | SIZE | REQ | LOCALITY | -------------------------------- | ---------------------------------------------------- | ---------- | ---------- | ----- | ----- | ---------- | 192.168.1.75,16020,1637588865433 | emp,,1637588876835.d6b9eaddeedbf2342c8e834a1151cc2a. | | | 1.0MB | 1 | 1.0 | 1 rows Took 0.0175 seconds {code} > TableSplit returns false size under 1MB > --------------------------------------- > > Key: HBASE-26340 > URL: https://issues.apache.org/jira/browse/HBASE-26340 > Project: HBase > Issue Type: Bug > Reporter: Norbert Kalmár > Assignee: Norbert Kalmár > Priority: Major > > We calculate region size in the mapreduce package by getting the size in MB > first and multiplying: > https://github.com/apache/hbase/blob/39a20c528e2bf27cedf12734dbdb1b7b1e538076/hbase-mapreduce/src/main/java/org/apache/hadoop/hbase/mapreduce/RegionSizeCalculator.java#L87 > This will give a size of 0 until at least 1MB is reached. (And it will have > an unwanted rounding affect as well). > Spark for example can be tuned to do some performance tuning by eliminating > the 0 sized regions. This will eliminate any small regions which are not > actually empty. The hadoop interface states the size is returned in bytes, > and while this is true do to the multiplication, we multiply by 0 until 1MB > is reached. I'm not sure why we get the size in MB units and not in bytes > straight up. -- This message was sent by Atlassian Jira (v8.20.1#820001)