[HBASE-18500] Performance issue: Don't use BufferedMutator for HTable's put method

• Type: Improvement
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Fix Version/s: 2.0.0
• Hadoop Flags: Incompatible change
• Release Note: Remove the deprecated method get/setWriteBufferSize from Table and remove writeBufferSize from TableBuilder. Remove the BufferedMutatorImpl from HTable.
• Assignee: Guanghao Zhang
• Reporter: Guanghao Zhang
• Created: 02/Aug/17 05:50
• Updated: 06/Apr/18 04:58
• Resolved: 11/Aug/17 08:43

Description

Copied the test result from HBASE-17994.
Run start-hbase.sh in my local computer and use the default config to test with PE tool.

./bin/hbase org.apache.hadoop.hbase.PerformanceEvaluation --rows=100000 --nomapred --autoFlush=True randomWrite 1
./bin/hbase org.apache.hadoop.hbase.PerformanceEvaluation --rows=100000 --nomapred --autoFlush=True asyncRandomWrite 1

Mean latency test result.

	Test1	Test2	Test3	Test4	Test5
randomWrite	164.39	161.22	164.78	140.61	151.69
asyncRandomWrite	122.29	125.58	122.23	113.18	123.02

50th latency test result.

	Test1	Test2	Test3	Test4	Test5
randomWrite	130.00	125.00	123.00	112.00	121.00
asyncRandomWrite	95.00	97.00	95.00	88.00	95.00

99th latency test result.

	Test1	Test2	Test3	Test4	Test5
randomWrite	600.00	600.00	650.00	404.00	425.00
asyncRandomWrite	339.00	327.00	297.00	311.00	318.00

In our internal 0.98 branch, the PE test result shows the async write has the almost same latency with the blocking write. But for master branch, the result shows the async write has better latency than the blocking client. Take a look about the code, I thought the difference is the BufferedMutator. For master branch, HTable don't have a write buffer and all write request will be flushed directly. And user can use BufferedMutator when user want to perform client-side buffering of writes. For the performance issue (autoFlush=True), I thought we can use rpc caller directly in HTable's put method. Thanks.

Review: https://reviews.apache.org/r/61454/

Copy the comments from chia7712. Remove the BufferdMutator brings four benefits.
1. correct the metrics (see HBASE-18476)
2. make HTable thread-safe (see HBASE-17368)
3. reduce the latency
4. get rid of some deprecated methods in Table

Issue Links

• is related to: HBASE-17368 Reduce contention on BufferedMutatorImpl#flush

Sub-Tasks

• Cleanup the docs saying "HTable use write buffer": Resolved