> We lost 78.5 seconds somewhere. From the sound of the disk drives, I'd
> say we lost it to seeking, which physical readahead with a large cache
> would be able to largely eliminate in this case.
So you only get a large % of your 78.5 seconds back
from pure physical readahead if the files and metadata
are (a) ordered in monitonically increasing disk
locations w.r.t. access order, and (b) physically close (no time
wasted reading in irrelevant data), or you apply some
form of clairvoyance patch :-) An alternative benchmark
would be do dd the /entire/ disk into RAM, then
run your diff on that, and I bet you get the
opposite result.
More serious point: if we retain readahead at a
logical level, you get it for non-physical
files too (e.g. NFS) - I presume this is
the intention. If so, what advantage does
additional physical readahead give you,
given physical ordering is surely no better
(and probably worse than) logical ordering.
-- Alex Bligh - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/