In going from 5000x5000 to 10000x10000 Magma's time increases by a factor of less than 4. That is impossible. Strassen will never help us there. They must be doing something else. Probably something clever.
Bill. On 17 May, 02:08, Bill Hart <[EMAIL PROTECTED]> wrote: > That may not be necessary. It may only be necessary to know the size > of the L1 and L2 caches and then it is probably possible to figure out > the optimal values. Probably it is something like 2^k rows of B must > fit in half the L1 cache and BLOCK rows of A must fit in half the L1 > cache. > > I'm not sure what happens with the L2 cache yet, since we haven't > really solved that issue yet. > > Bill. > > On 17 May, 01:53, mabshoff <[EMAIL PROTECTED]> wrote: > > > On May 17, 2:48 am, Bill Hart <[EMAIL PROTECTED]> wrote: > > > > Hi Martin, > > > > That spike is wierd. Basically I got closer to 2x the time of Magma > > > for 16384x16384, but you need different parameters than for > > > 10000x10000 or 20000x20000 since the size of the M4R matrices will be > > > different than in either of those cases. > > > > For 16384x16384 my notes say that I used k = 6 and BLOCK = 256. One > > > might also need to fiddle with the cutoff. I think I used 3600, but at > > > various times I had the cutoff set lower. > > > > It would be awfully surprising if there wasn't a set of parameters > > > that dropped this time right down. It is possible that the L1 cache is > > > smaller on sage.math than on the Opteron I was using. Perhaps my > > > parameters don't apply on that machine. > > > Shouldn't we use an ATLAS like tuning process? [obviously not that > > heavy timewise]. It will likely find good default values. > > > > Bill. > > > <SNIP> --~--~---------~--~----~------------~-------~--~----~ To post to this group, send email to sage-devel@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sage-devel URLs: http://www.sagemath.org -~----------~----~----~----~------~----~------~--~---
