On Mar 30, 2007, at 8:14 AM, Dimitri wrote:
>> You are right in that the page size constraint is lifted in that
>> directio cuts out the VM filesystem cache. However, the Solaris
>> kernel still issues io ops in terms of its logical block size (which
>> we have at the default 8K). It can issue io ops for fragments as
>> small as 1/8th of the block size, but Postgres issues its io requests
>> in terms of the block size which means that io ops from Postgres will
>> be in 8K chunks which is exactly what we see when we look at our
>> system io stats. In fact, if any io request is made that isn't a
>> multiple of 512 bytes (the disk sector size), the file system
>> switches back to the buffered io.
> Oh, yes, of course! yes, you still need to respect multiple of 512
> bytes block size on read and write - sorry, I was tired :)
> Then it's seems to be true - default XLOG block size is 8K, means for
> every even small auto-committed transaction we should write 8K?... Is
> there any reason to use so big default block size?...
> Probably it may be a good idea to put it as 'initdb' parameter? and
> have such value per database server?
I believe it's because that is a pretty normal Unix kernal block size
and you want the two to match.
erik jones <erik(at)myemma(dot)com>
In response to
pgsql-performance by date
|Next:||From: Dimitri||Date: 2007-03-30 14:25:16|
|Subject: Re: Shared buffers, db transactions commited, and write IO on Solaris|
|Previous:||From: Matteo Beccati||Date: 2007-03-30 13:36:26|
|Subject: Re: Wrong plan sequential scan instead of an index one|