Re: [HACKERS] wait events for disk I/O - Mailing list pgsql-hackers

From Rajkumar Raghuwanshi
Subject Re: [HACKERS] wait events for disk I/O
Date
Msg-id CAKcux6kHb+zAYhed8Sca9wLn-3fS2B3whdghrMVqkfgPW-5D=w@mail.gmail.com
Whole thread Raw
In response to [HACKERS] wait events for disk I/O  (Rushabh Lathia <rushabh.lathia@gmail.com>)
Responses Re: [HACKERS] wait events for disk I/O  (Rushabh Lathia <rushabh.lathia@gmail.com>)
List pgsql-hackers
On Wed, Mar 8, 2017 at 4:50 PM, Rushabh Lathia <rushabh.lathia@gmail.com> wrote:
I am attaching  another version of the patch, as I found stupid mistake
in the earlier version of patch, where I missed to initialize initial value to
WaitEventIO enum. Also earlier version was not getting cleanly apply on
the current version of sources.

I have applied attached patch, set shared_buffers to 128kB and ran pgbench, I am able to see below distinct IO events.

--with ./pgbench -i -s 500 postgres
application_name    wait_event_type    wait_event              query
 pgbench                  IO                          ExtendDataBlock   copy pgbench_account
 pgbench                  IO                          WriteXLog              copy pgbench_account
 pgbench                  IO                          WriteDataBlock      copy pgbench_account
 pgbench                  IO                          ReadDataBlock      vacuum analyze pgben
 pgbench                  IO                          ReadBuffile            alter table pgbench_

--with ./pgbench -T 600 postgres (readwrite)
application_name    wait_event_type    wait_event              query
 pgbench                 IO                           ReadDataBlock      UPDATE pgbench_accou
 pgbench                 IO                           WriteDataBlock      UPDATE pgbench_telle
                                IO                           SyncDataBlock    
 pgbench                 IO                           SyncDataBlock      UPDATE pgbench_telle
                                IO                           SyncDataBlock      autovacuum: VACUUM A
 pgbench                 IO                           WriteXLog              END;
 pgbench                 IO                           ExtendDataBlock   copy pgbench_account

--with ./pgbench -T 600 -S postgres (select only)
application_name    wait_event_type    wait_event              query
 pgbench                 IO                           ReadDataBlock      SELECT abalance FROM

Attached excel with all IO event values.

Thanks & Regards,
Rajkumar Raghuwanshi
QMG, EnterpriseDB Corporation


Attachment

pgsql-hackers by date:

Previous
From: David Rowley
Date:
Subject: Re: [HACKERS] Foreign Join pushdowns not working properly for outer joins
Next
From: Robert Haas
Date:
Subject: Re: [HACKERS] Parallel bitmap heap scan