Re: [REVIEW] Re: Compression of full-page-writes

Hello,

>Maybe. Let's get the basic patch done first; then we can argue about that

Please find attached patch to compress FPW using pglz compression.
All backup blocks in WAL record are compressed at once before inserting it
into WAL buffers . Full_page_writes GUC has been modified to accept three
values
1. On
2. Compress
3. Off
FPW are compressed when full_page_writes is set to compress. FPW generated
forcibly during online backup even when full_page_writes is off are also
compressed. When full_page_writes is set on FPW are not compressed.
Benckmark:
Server Specification:
Processors:Intel® Xeon ® Processor E5-2650 (2 GHz, 8C/16T, 20 MB) * 2 nos
RAM: 32GB
Disk : HDD 450GB 10K Hot Plug 2.5-inch SAS HDD * 8 nos
1 x 450 GB SAS HDD, 2.5-inch, 6Gb/s, 10,000 rpm
Checkpoint segments: 1024
Checkpoint timeout: 5 mins
pgbench -c 64 -j 64 -r -T 900 -M prepared
Scale factor: 1000

WAL generated (MB) Throughput
(tps) Latency(ms)
On 9235.43
979.03 65.36
Compress(pglz) 6518.68
1072.34 59.66
Off 501.04 1135.17
56.34

The results show around 30 percent decrease in WAL volume due to
compression of FPW.
compress_fpw_v1.patch
<http://postgresql.1045698.n5.nabble.com/file/n5819645/compress_fpw_v1.patch>

--
View this message in context: http://postgresql.1045698.n5.nabble.com/Compression-of-full-page-writes-tp5769039p5819645.html
Sent from the PostgreSQL - hackers mailing list archive at Nabble.com.