[HBASE-10201] Port 'Make flush decisions per column family' to trunk

• Type: Improvement
• Status: Closed
• Priority: Major
• Resolution: Fixed
• Fix Version/s: 1.1.0, 2.0.0
• Component/s: wal
• Hadoop Flags: Reviewed

Release Note

Adds new flushing policy mechanism. Default, org.apache.hadoop.hbase.regionserver.FlushLargeStoresPolicy, will try to avoid flushing out the small column families in a region, those whose memstores are < hbase.hregion.percolumnfamilyflush.size.lower.bound. To restore the old behavior of flushes writing out all column families, set hbase.regionserver.flush.policy to org.apache.hadoop.hbase.regionserver.FlushAllStoresPolicy either in hbase-default.xml or on a per-table basis by setting the policy to use with HTableDescriptor.getFlushPolicyClassName().

• Assignee: Duo Zhang
• Reporter: Ted Yu
• Created: 18/Dec/13 19:08
• Updated: 19/Jul/15 12:33
• Resolved: 16/Dec/14 20:01

Description

Currently the flush decision is made using the aggregate size of all column families. When large and small column families co-exist, this causes many small flushes of the smaller CF. We need to make per-CF flush decisions.

Issue Links

• relates to: HBASE-13408 HBase In-Memory Memstore Compaction
• relates to: HBASE-3149 Make flush decisions per column family
• relates to: HBASE-12405 WAL accounting by Store
• links to: reviewboard