Thanks for the info.
[snip]
> > So, I am wondering whether people in the numerical
> > analysis community have started to think in this
> > direction and what they have come up with so far
> > (if anything).
>
> I don't think that the numerical analysis community has turned their
> attention much to map-reduce yet.
One other major consumer of large, sparse matrix
operations are people dealing with partial differential
equations. And the Linear Programming/Optimization
people also have algorithmically non-trivial, large
problems, that might (just might) be amenable to M/R.
On the other hand, I would not expect physics
simulations to work well - the communication
overhead will kill you (it did not work very well on
the Cray, back in the day, and the problems have
not changed.)
>
> > This is not a question about Mahout, and I admit that
> > it is therefore a little OT for this list, on the other hand,
> > I would assume that the members of this list would be
> > the first people to know about progress in this area...
>
> It is entirely on-topic. The biggest news I know of is the stochastic
> decomposition stuff that I mentioned before.
I see. Thanks again for the info.
Best,
Ph.
