From: | Ants Aasma <ants(dot)aasma(at)eesti(dot)ee> |
---|---|
To: | Tomas Vondra <tomas(dot)vondra(at)2ndquadrant(dot)com> |
Cc: | Konstantin Knizhnik <k(dot)knizhnik(at)postgrespro(dot)ru>, Andres Freund <andres(at)anarazel(dot)de>, Robert Haas <robertmhaas(at)gmail(dot)com>, Thomas Munro <thomas(dot)munro(at)enterprisedb(dot)com>, PostgreSQL-development <pgsql-hackers(at)postgresql(dot)org>, seanc(at)joyent(dot)com |
Subject: | Re: WAL prefetch |
Date: | 2018-06-19 13:57:25 |
Message-ID: | CA+CSw_tsQ2i_0Vpv8gg7qoHCPYXmU+HKCU1jakLA_EBwktCfbw@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
On Tue, Jun 19, 2018 at 4:04 PM Tomas Vondra <tomas(dot)vondra(at)2ndquadrant(dot)com>
wrote:
> Right. My point is that while spawning bgworkers probably helps, I don't
> expect it to be enough to fill the I/O queues on modern storage systems.
> Even if you start say 16 prefetch bgworkers, that's not going to be
> enough for large arrays or SSDs. Those typically need way more than 16
> requests in the queue.
>
> Consider for example [1] from 2014 where Merlin reported how S3500
> (Intel SATA SSD) behaves with different effective_io_concurrency values:
>
> [1]
>
> https://www.postgresql.org/message-id/CAHyXU0yiVvfQAnR9cyH=HWh1WbLRsioe=mzRJTHwtr=2azsTdQ@mail.gmail.com
>
> Clearly, you need to prefetch 32/64 blocks or so. Consider you may have
> multiple such devices in a single RAID array, and that this device is
> from 2014 (and newer flash devices likely need even deeper queues).'
>
For reference, a typical datacenter SSD needs a queue depth of 128 to
saturate a single device. [1] Multiply that appropriately for RAID arrays.
Regards,
Ants Aasma
From | Date | Subject | |
---|---|---|---|
Next Message | Robert Haas | 2018-06-19 14:15:33 | Re: server crashed with TRAP: FailedAssertion("!(!parallel_aware || pathnode->path.parallel_safe)" |
Previous Message | Jeremy Finzel | 2018-06-19 13:50:17 | Re: found xmin from before relfrozenxid on pg_catalog.pg_authid |