Re: Setting behavior for clustered IO

[Samuel Thibault]

Samuel Thibault, le Tue 03 Apr 2012 00:05:39 +0200, a écrit :
> Sergio López, le Tue 03 Apr 2012 00:00:02 +0200, a écrit :
> > 2012/4/2 Samuel Thibault <samuel.thibault@gnu.org>:
> > > Sergio Lopez, le Mon 02 Apr 2012 17:09:08 +0200, a écrit :
> > >> El Mon, 02 Apr 2012 00:23:03 +0300
> > >> Maksym Planeta <mcsim.planeta@gmail.com> escribió:
> > >> > This function allows user advise the kernel about how to handle paging
> > >> > input/output in specified memory range. There are several behaviors,
> > >> > like RANDOM, NORMAL, SEQUENTIAL, WILLNEED and DONTNEED. From the page
> > >> > fault handler's point of view these behaviors differ only in size of
> > >> > memory chunk that will be read ahead.
> > >>
> > >> I don't think the kernel should be the one to be advised, but the
> > >> filesystem translators. These are the ones who really know current and
> > >> future (as they control most of the operations) state of the object,
> > >
> > > Do they really?  We discussed about it with neal a long time ago, and we
> > > believed that Mach was at a better position, because it knows about
> > > each and every mapping. Say for instance that two processes map the same
> > > file, and access it concurrently, but in a different way. AIUI, the
> > > translators will get data requests without indication of what mapping is
> > > pulling it, and thus no correlation between them, thus seemingly
> > > random.
> 
> I realize that I was talking for read-ahead, which has not much to do
> with cluster size.

That being said, madvise(SEQUENTIAL) should not only set the cluster
size, but more importantly, the readahead. The whole point of readahead
is to pipeline actual disk reads and processing, otherwise we don't do
both in parallel, be it clusterized or not.

Samuel