[ https://issues.apache.org/jira/browse/HBASE-5074?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Zhihong Yu updated HBASE-5074: ------------------------------ Comment: was deleted (was: tedyu has commented on the revision "[jira] [HBASE-5074] Support checksums in HBase block cache". INLINE COMMENTS src/main/java/org/apache/hadoop/hbase/io/hfile/HFile.java:425 This cast is not safe. See https://builds.apache.org/job/PreCommit-HBASE-Build/907//testReport/org.apache.hadoop.hbase.mapreduce/TestLoadIncrementalHFiles/testSimpleLoad/: Caused by: java.lang.ClassCastException: org.apache.hadoop.hdfs.DistributedFileSystem cannot be cast to org.apache.hadoop.hbase.util.HFileSystem at org.apache.hadoop.hbase.io.hfile.HFile.createReaderWithEncoding(HFile.java:425) at org.apache.hadoop.hbase.io.hfile.HFile.createReader(HFile.java:433) at org.apache.hadoop.hbase.mapreduce.LoadIncrementalHFiles.groupOrSplit(LoadIncrementalHFiles.java:407) at org.apache.hadoop.hbase.mapreduce.LoadIncrementalHFiles$2.call(LoadIncrementalHFiles.java:328) at org.apache.hadoop.hbase.mapreduce.LoadIncrementalHFiles$2.call(LoadIncrementalHFiles.java:326) src/main/java/org/apache/hadoop/hbase/io/hfile/HFile.java:160 Should we default to CRC32C ? src/main/java/org/apache/hadoop/hbase/util/ChecksumFactory.java:2 No year is needed. src/main/java/org/apache/hadoop/hbase/util/ChecksumFactory.java:59 Shall we name this variable ctor ? Similar comment applies to other meth variables in this patch. REVISION DETAIL https://reviews.facebook.net/D1521 ) > support checksums in HBase block cache > -------------------------------------- > > Key: HBASE-5074 > URL: https://issues.apache.org/jira/browse/HBASE-5074 > Project: HBase > Issue Type: Improvement > Components: regionserver > Reporter: dhruba borthakur > Assignee: dhruba borthakur > Attachments: D1521.1.patch, D1521.1.patch, D1521.2.patch, > D1521.2.patch, D1521.3.patch, D1521.3.patch > > > The current implementation of HDFS stores the data in one block file and the > metadata(checksum) in another block file. This means that every read into the > HBase block cache actually consumes two disk iops, one to the datafile and > one to the checksum file. This is a major problem for scaling HBase, because > HBase is usually bottlenecked on the number of random disk iops that the > storage-hardware offers. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira