Horton
Menu

Re: increasing the default WAL segment size

Michael Paquier <michael.paquier@gmail.com>

View thread
From: Michael Paquier <michael.paquier@gmail.com>
To: Robert Haas <robertmhaas@gmail.com>
Cc: Simon Riggs <simon@2ndquadrant.com>, Amit Kapila <amit.kapila16@gmail.com>, Jim Nasby <Jim.Nasby@bluetreble.com>, PostgreSQL-development <pgsql-hackers@postgresql.org>, Beena Emerson <memissemerson@gmail.com>
Date: 2017-01-04T23:45:03Z
Lists: pgsql-hackers 
On Thu, Jan 5, 2017 at 12:33 AM, Robert Haas <robertmhaas@gmail.com> wrote:
> On Wed, Jan 4, 2017 at 9:47 AM, Simon Riggs <simon@2ndquadrant.com> wrote:
>> On 4 January 2017 at 13:57, Robert Haas <robertmhaas@gmail.com> wrote:
>>> On Wed, Jan 4, 2017 at 3:05 AM, Simon Riggs <simon@2ndquadrant.com> wrote:
>>>> Strange response. Nothing has been assumed. I asked for tests and you
>>>> provided measurements.
>>>
>>> Sure, of zero-filling a file with dd.  But I also pointed out that in
>>> a real PostgreSQL cluster, the change could actually *reduce* latency.
>>
>> I think we are talking at cross purposes. We agree that the main
>> change is useful, but it causes another problem which I can't see how
>> you can characterize as reduced latency, based upon your own
>> measurements.
>
> Zero-filling files will take longer if the files are bigger.  That
> will increase latency.  But we will also have fewer forced
> end-of-segment syncs.  That will reduce latency.  Which effect is
> bigger?

It depends on if the environment is CPU-bounded or I/O bounded. If CPU
is at its limit, zero-filling takes time. If that's the I/O, fsync()
would take longer to complete.
-- 
Michael

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →

  1. Make WAL segment size configurable at initdb time.

  2. Perform only one ReadControlFile() during startup.

  3. Introduce BYTES unit for GUCs.

  4. Remove useless duplicate inclusions of system header files.

  5. Refactor other replication commands to use DestRemoteSimple.

  6. Add a SHOW command to the replication command language.

  7. Add a new DestReceiver for printing tuples without catalog access.

  8. Support fls().

  9. Extend yesterday's patch making BLCKSZ and RELSEG_SIZE configurable to also

  10. Commit the reasonably uncontroversial parts of J.R. Nield's PITR patch, to

  11. XLOG (also known as WAL -:)) Bootstrap/Startup/Shutdown.

  12. Transaction log manager core code.