Re: wait events for disk I/O

Thanks Rajkumar for performing tests on this patch.

Yes, I also noticed similar results in my testing. Additionally sometime I
also
noticed ReadSLRUPage event on my system.

I also run the reindex database command and I notices below IO events.

SyncImmedRelation,
WriteDataBlock
WriteBuffile,
WriteXLog,
ReadDataBlock

On Wed, Mar 8, 2017 at 6:41 PM, Rajkumar Raghuwanshi <
rajkumar(dot)raghuwanshi(at)enterprisedb(dot)com> wrote:

> On Wed, Mar 8, 2017 at 4:50 PM, Rushabh Lathia <rushabh(dot)lathia(at)gmail(dot)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
>
>
>

--
Rushabh Lathia
www.EnterpriseDB.com