Обсуждение: TODO: Determine optimal fdatasync/fsync, O_SYNC/O_DSYNC options

Поиск
Список
Период
Сортировка

TODO: Determine optimal fdatasync/fsync, O_SYNC/O_DSYNC options

От
"Joshua D. Drake"
Дата:
Doesn't this exist in:

src/tools/fsync?

Do we just need to make it more user friendly?

Joshua D. Drake
-- 
   === The PostgreSQL Company: Command Prompt, Inc. ===
Sales/Support: +1.503.667.4564 || 24x7/Emergency: +1.800.492.2240   Providing the most comprehensive  PostgreSQL
solutionssince 1997             http://www.commandprompt.com/
 




Re: TODO: Determine optimal fdatasync/fsync, O_SYNC/O_DSYNC options

От
Tom Lane
Дата:
"Joshua D. Drake" <jd@commandprompt.com> writes:
> Doesn't this exist in:
> src/tools/fsync?

No, that only tells you if the various alternatives *work*.  Not which
one performs best.
        regards, tom lane


Re: TODO: Determine optimal fdatasync/fsync, O_SYNC/O_DSYNC options

От
Josh Berkus
Дата:
Tom, JD,

> No, that only tells you if the various alternatives *work*.  Not which
> one performs best.

"Performs best" is pretty elusive.   On Solaris 10, for example, fdatasync 
does best on pgbench on x86, but opendatasync does best on DBT2 on Sparc.  
I'm not certain we can really determine this with any accuracy.

-- 
--Josh

Josh Berkus
PostgreSQL @ Sun
San Francisco


Re: TODO: Determine optimal fdatasync/fsync, O_SYNC/O_DSYNC options

От
Tom Lane
Дата:
Josh Berkus <josh@agliodbs.com> writes:
> "Performs best" is pretty elusive.   On Solaris 10, for example, fdatasync 
> does best on pgbench on x86, but opendatasync does best on DBT2 on Sparc.  
> I'm not certain we can really determine this with any accuracy.

Yeah, if it were easy we've have done it already :-(.  One thing that
would be useful though is to try to gather datapoints like the above
for a range of different operating systems.  Even if we could tell
people "here are the two or three settings worth trying for your OS",
we'd be ahead of the game.
        regards, tom lane