Re: Anyone for SSDs? - Mailing list pgsql-hackers

From Robert Haas
Subject Re: Anyone for SSDs?
Date
Msg-id AANLkTinHTw_JCxrfVUSiVQt4Wc28VyewSSE5YY_VAT9X@mail.gmail.com
Whole thread Raw
In response to Anyone for SSDs?  (Vaibhav Kaushal <vaibhavkaushal123@gmail.com>)
Responses Re: Anyone for SSDs?  (Tom Lane <tgl@sss.pgh.pa.us>)
Re: Anyone for SSDs?  (Vaibhav Kaushal <vaibhavkaushal123@gmail.com>)
List pgsql-hackers
On Fri, Dec 10, 2010 at 1:39 AM, Vaibhav Kaushal
<vaibhavkaushal123@gmail.com> wrote:
> Most of you already know I am new to this list and newer to any OSS
> development. However, while browsing the source code (of 9.0.1) I find
> that there is only one way to store relations on disk - the magnetic
> disk.
>
> This came suddenly in my mind so I am asking the experts here.
>
> Considering the fact that SSDs will be common (at least for the
> enterprise) in the coming years because of (of course you know the
> reason) their less seek time and higher transfer rates per second, is
> there someone trying for a ssd.c? In almost all cases even using md.c,
> the kernel will handle it effectively but would it not be better that we
> are well prepared to ask kernel for more?
>
> Or has such an attempt already begun?

Questions about using SSDs with PostgreSQL would be more appropriate
on pgsql-performance, rather than here.  If you search, you'll find
that the topic has been covered extensively in the archives.

But as far as the code goes, there doesn't seem to be any reason why
SSDs would require any changes to md.c, or an alternate
implementation.  The interface the operating system presents is the
same.

-- 
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company


pgsql-hackers by date:

Previous
From: Robert Haas
Date:
Subject: Re: initdb failure with Postgres 8.4.4
Next
From: Andrew Dunstan
Date:
Subject: Re: initdb failure with Postgres 8.4.4