Bulk Operation

HBASE-20188 Size of in-memory compaction thread pool should be configurable

HBASE-20188 Most of KeyValueUtil#length can be replaced by cell#getSerializedSize for better performance because the latter one has been optimized

HBASE-20188 Remove all the CSLM#size operation in our memstore because it's an quite time consuming.

HBASE-20188 MemStoreLABImp::copyIntoCell uses 7% CPU when writing

HBASE-20188 Save recalculating families in a WALEdit batch of Cells

HBASE-20188 ServerRpcConnection logging cleanup

HBASE-20188 Unsafe access cleanup

HBASE-20188 extra cloneFamily() in Mutation.add(Cell)

HBASE-20188 SegmentScanner does over-comparing when one flushing

HBASE-20188 Tighter ByteBufferKeyValue Cell Comparator; part 2

HBASE-20188 Tighter ByteBufferKeyValue Cell Comparator

HBASE-20188 Better heap utilization for IMC with MSLABs

HBASE-20188 [PERFORMANCE] Flushing is 2x slower in hbase2.

HBASE-20188 [IHC] Workloadx

HBASE-20188 [DOC] Add note on perf to upgrade section

HBASE-20188 IMC Default Parameters for 2.0.0

HBASE-20188 Update the doc on how to setup shortcircuit reads; its stale

HBASE-20188 [locking] Write-time worst offenders