Postgres Pro
Downloads
Home   >   mailing lists

Re: [HACKERS] Should we cacheline align PGXACT? - Mailing list pgsql-hackers

From Simon Riggs
Subject Re: [HACKERS] Should we cacheline align PGXACT?
Date
Msg-id CANP8+j+=nUS+WgX_kcJwNtVqHuVJgYeOyW9-MgDd_=j6MQyCyQ@mail.gmail.com
Whole thread Raw
In response to Re: [HACKERS] Should we cacheline align PGXACT?  (Ashutosh Sharma <ashu.coek88@gmail.com>)
Responses Re: [HACKERS] Should we cacheline align PGXACT?  (Ashutosh Sharma <ashu.coek88@gmail.com>)
List pgsql-hackers
Tree view
On 28 February 2017 at 11:34, Ashutosh Sharma <ashu.coek88@gmail.com> wrote:
 
So, Here are the pgbench results I got with 'reduce_pgxact_access_AtEOXact.v2.patch' on a read-write workload.

Thanks for performing a test.

I see a low yet noticeable performance gain across the board on that workload.

That is quite surprising to see a gain on that workload. The main workload we have been discussing was the full read-only test (-S). For that case the effect should be much more noticeable based upon Andres' earlier comments.

Would it be possible to re-run the test using only the -S workload? Thanks very much.
 

pgbench settings:
pgbench -i -s 300 postgres
pgbench -M prepared -c $thread -j $thread -T $time_for_reading  postgres

where, time_for_reading = 30mins

non default GUC param
shared_buffers=8GB
max_connections=300

pg_wal is located in SSD.

CLIENT COUNTTPS (HEAD)TPS (PATCH)% IMPROVEMENT
4258826010.5023183926
8509450980.0785237534
1610294103070.1262871576
3219779198150.182011224
6427908283461.569442454
7227823284162.131330194
12828455286180.5728342998
18026739268790.5235797898
19627820279630.5140186916
25628763289690.7161978931


Also, Excel sheet (results-readwrite-300-SF.xlsx) containing the results for all the 3 runs is attached.



--
Simon Riggs                http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services

pgsql-hackers by date:

Previous
From: Dmitry Dolgov
Date:
Subject: [HACKERS] [PATCH] Generic type subscripting
Next
From: Stephen Frost
Date:
Subject: [HACKERS] New Committer - Andrew Gierth