Обсуждение: Filesystem optimisation for postgresql tables and WAL logs on linux.

Hi,


I have got there SCSIs HDDs for the postgresql Box
i plan to install OS on 1st , 2nd for tables and indexes and third
for pg_xlog.

Bruce Momjian's "H/W Perf tuning" on
http://www.ca.postgresql.org/docs/momjian/hw_performance/node11.html
recommends 8k blocksize for the filesystem which is same as the page size.

On linux ext2 fs (on my server) 4096 is the default blocak size and 8192 bytes
is not supported.

but man mkfs.ext2 on linux mentions an option -T which is:
-T fs-type
              Specify  how  the filesystem is going to be used, so that mke2fs can chose optimal filesystem parameters
forthat 
              use.  The supported filesystem types are:
                  news        one inode per 4kb block
                  largefile   one inode per megabyte
                  largefile4  one inode per 4 megabytes

is the above relevent as far as optimisation for filesystem for tables is concerned?

Also for the pg_xlog drive is a particular block size more favourable then others?


regds
mallah.






--
Rajesh Kumar Mallah,
Project Manager (Development)
Infocom Network Limited, New Delhi
phone: +91(11)6152172 (221) (L) ,9811255597 (M)

Visit http://www.trade-india.com ,
India's Leading B2B eMarketplace.

I don't remember saying they should match, it is just that it is nice if
it does, but I don't think it would make any major difference in
performance.  In fact, some use 32k pages sizes, and get a performance
boost, and clearly don't have 32k file system blocks.

---------------------------------------------------------------------------

Rajesh Kumar Mallah. wrote:
>
> Hi,
>
>
> I have got there SCSIs HDDs for the postgresql Box
> i plan to install OS on 1st , 2nd for tables and indexes and third
> for pg_xlog.
>
> Bruce Momjian's "H/W Perf tuning" on
> http://www.ca.postgresql.org/docs/momjian/hw_performance/node11.html
> recommends 8k blocksize for the filesystem which is same as the page size.
>
> On linux ext2 fs (on my server) 4096 is the default blocak size and 8192 bytes
> is not supported.
>
> but man mkfs.ext2 on linux mentions an option -T which is:
> -T fs-type
>               Specify  how  the filesystem is going to be used, so that mke2fs can chose optimal filesystem
parametersfor that 
>               use.  The supported filesystem types are:
>                   news        one inode per 4kb block
>                   largefile   one inode per megabyte
>                   largefile4  one inode per 4 megabytes
>
> is the above relevent as far as optimisation for filesystem for tables is concerned?
>
> Also for the pg_xlog drive is a particular block size more favourable then others?
>
>
> regds
> mallah.
>
>
>
>
>
>
> --
> Rajesh Kumar Mallah,
> Project Manager (Development)
> Infocom Network Limited, New Delhi
> phone: +91(11)6152172 (221) (L) ,9811255597 (M)
>
> Visit http://www.trade-india.com ,
> India's Leading B2B eMarketplace.
>
>
>
> ---------------------------(end of broadcast)---------------------------
> TIP 1: subscribe and unsubscribe commands go to majordomo@postgresql.org
>

--
  Bruce Momjian                        |  http://candle.pha.pa.us
  pgman@candle.pha.pa.us               |  (610) 359-1001
  +  If your life is a hard drive,     |  13 Roberts Road
  +  Christ can be your backup.        |  Newtown Square, Pennsylvania 19073