On Thu, 2003-03-27 at 21:42, David Boyes wrote: > > But this does not simulate the same effect. > > Field testing seems to indicate that there is a noticeable benefit to > this approach. > Is it equivalent to PAV? Maybe not. Perhaps I don't understand PAV well > enough.
I have done some work on this, both during the residency and back at home. I may well be 6 months serious work away from understanding disk I/O with Linux on VM.I think we have some 6-7 layers of cache and blocking, each with their own schemes. It is complicated. There's so many non-linear things in the curve that 'field test' will not tell you anything (well, except that it works for that case that you tested). But don't think you can extrapolate or even interpolate the results.
