Postgres Pro
Downloads
Home   >   mailing lists

Should we cacheline align PGXACT? - Mailing list pgsql-hackers

From Alexander Korotkov
Subject Should we cacheline align PGXACT?
Date
Msg-id CAPpHfdtJY4zOEDsjad6J5AyZMqZcv6gSY9AkKpA7qN3jyQ2+1Q@mail.gmail.com
Whole thread Raw
Responses Re: Should we cacheline align PGXACT?  (Amit Kapila <amit.kapila16@gmail.com>)
Re: Should we cacheline align PGXACT?  (Tom Lane <tgl@sss.pgh.pa.us>)
Re: Should we cacheline align PGXACT?  (Robert Haas <robertmhaas@gmail.com>)
List pgsql-hackers
Tree view
Hackers,

originally this idea was proposed by Andres Freund while experimenting with lockfree Pin/UnpinBuffer [1].
The patch is attached as well as results of pgbench -S on 72-cores machine.  As before it shows huge benefit in this case.
For sure, we should validate that it doesn't cause performance regression in other cases.  At least we should test read-write and smaller machines.
Any other ideas?


------
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
Attachment

pgsql-hackers by date:

Previous
From: Tatsuo Ishii
Date:
Subject: Re: Slowness of extended protocol
Next
From: Tatsuo Ishii
Date:
Subject: Exporting more function in libpq