Do you see this getting put into the ACTS toolkit (http://acts.nersc.gov/tools.html)? I see some on your project team are at Argonne, which would seem like a natural fit.

Elemental is available via PETSc (http://mcs.anl.gov/petsc), which is part of ACTS. We might be able to talk ACTS into giving Jack a session at the tutorial session next year. (Disclaimer: I'm a PETSc developer and helped with the Elemental interface.)

Honestly, I doubt it. Getting added to such a list has both political and merit-based components.

Interesting. I had the impression it was just meant to be a repository for DOE-related numerical tools.

That's a shame, because having collections of useful tools like that makes it a lot easier to keep track of options.

Given that a few dozen US-based students attend ACTS every year, I hardly see how it is a game-changer w.r.t. the widespread acceptance of Elemental. I think the internet is doing just fine so far.

what is the range of matrix sizes that would enjoy a speedup over ScaLAPACK? let's assume a 8 or 16-core cluster with N nodes, so, a function of N.

There is unfortunately no good answer to this question, as it would completely depend upon the algorithm, MPI implementation, communication network, vendor-tuned BLAS library, etc., etc. However, I can safely say that Elemental is always competitive with ScaLAPACK in performance. The Elemental papers on my website (pick between the preprint or journal article), http://www.stanford.edu/~poulson, go into this in more detail, but keep in mind that the article was written a couple of years ago, when it was still a fledgling library, and so the tone is a bit more aggressive than necessary.

With that said, the main advantage of the library is its high-level of software-engineering, which tends to encourage the rapid development of new features. It is actively used within a large number of research projects.

Indeed, we didn't chose Elemental over ScaLAPACK; we used Elemental to replace LAPACK because it was impossible to refactor our code to use ScaLAPACK without absolutely ruining it forever. The ScaLAPACK interface is horrible and should only be used inside of a higher-level library interface that hides its ugly, ugly control structures.

From my perspective, Elemental is primarily about human performance. That it scales to thousands of nodes of Blue Gene and achieves performance on par with (and in some cases significantly better than) ScaLAPACK is really icing on the cake. Granted, I really like scalability and performance, but I will not ruin my code in the name of these things.

There's no need to be a Negative Nancy, especially when you're affiliated with the project you're promoting (without disclosing it).

I love you guys.