Re: [PLUG] swappiness and ssd

[Rich Freeman <r-plug@thefreemanclan.net>]

On Apr 16, 2015 10:42 AM, "Walt Mankowski" <waltman@pobox.com> wrote:
>
> On Thu, Apr 16, 2015 at 10:01:55AM -0400, K.S. Bhaskar wrote:
> > The default swappiness on Ubuntu systems is 60, and presumably takes into
> > consideration the relative performance of writes to and reads from the swap
> > file / partition.  The ratio of write performance to read performance would
> > be higher for SSDs than rotating media, since SSD reads are truly random
> > whereas writes are in pages (in rotating media, seek and latency times
> > would tend to apply to both).  Therefore, it seems to me that if swap is
> > located on SSD, we should increase swappiness, to encourage less frequently
> > used pages to be written out, since they can be brought back in very fast
> > when needed.  Does that seem reasonable or am I missing something?  Thanks.
>
> I'm not sure why you think that SSD reads are "truly random".  First,
> many reads will be buffered, so you'll still be reading in pages at a
> time instead of bytes at a time.  Also, keep in mind the kernel is in
> the way of all your reads.  It can cache frequently accessed data, and
> also reorder reads so that they're performed more efficiently.

Wouldn't caching the page file be a bit counterproductive?

Generally speaking what you say is true, but not so much in this case.

Rich

___________________________________________________________________________
Philadelphia Linux Users Group         --        http://www.phillylinux.org
Announcements - http://lists.phillylinux.org/mailman/listinfo/plug-announce
General Discussion  --   http://lists.phillylinux.org/mailman/listinfo/plug