From: | Konstantin Knizhnik <k(dot)knizhnik(at)postgrespro(dot)ru> |
---|---|
To: | Andres Freund <andres(at)anarazel(dot)de> |
Cc: | pgsql-hackers(at)lists(dot)postgresql(dot)org |
Subject: | Re: Improving connection scalability: GetSnapshotData() |
Date: | 2020-09-07 14:20:53 |
Message-ID: | 311c62b6-b584-1db8-26aa-07f38c296fdd@postgrespro.ru |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
On 06.09.2020 21:56, Andres Freund wrote:
>
> Hm, that is interesting / odd. If you record a profile with call graphs
> (e.g. --call-graph dwarf), where are all the LWLockAttemptLock calls
> comming from?
>
Attached.
> I assume the machine you're talking about is an 8 socket machine?
>
> What if you:
> a) start postgres and pgbench with numactl --interleave=all
> b) start postgres with numactl --interleave=0,1 --cpunodebind=0,1 --membind=0,1
> in case you have 4 sockets, or 0,1,2,3 in case you have 8 sockets?
>
TPS for -c 100
--interleave=all
1168910
--interleave=0,1
1232557
--interleave=0,1,2,3
1254271
--cpunodebind=0,1,2,3 --membind=0,1,2,3
1237237
--cpunodebind=0,1 --membind=0,1
1420211
--cpunodebind=0 --membind=0
1101203
>> And which pgbench database scale factor you have used?
> 200
>
> Another thing you could try is to run 2-4 pgench instances in different
> databases.
I tried to reinitialize database with scale 200 but there was no
significant improvement in performance.
Attachment | Content-Type | Size |
---|---|---|
pgbench.svg | image/svg+xml | 172.1 KB |
From | Date | Subject | |
---|---|---|---|
Next Message | Anastasia Lubennikova | 2020-09-07 14:23:58 | Re: 回复:how to create index concurrently on partitioned table |
Previous Message | Heikki Linnakangas | 2020-09-07 14:10:08 | Re: Yet another fast GiST build |