[gmx-developers] Re: Posit-enablement of GROMACS
theo at stillwater-sc.com
Sat Aug 4 20:12:48 CEST 2018
On Wed, Aug 1, 2018, 16:12 Mark Abraham <mark.j.abraham at gmail.com> wrote:
> It does sound like a generic hope, rather than one based on extensive
> understanding of how high performance MD codes work. :-) I don't mean to
> negative, but MD codes are probably not the low hanging fruit for
> demonstrating the benefits of this approach.
It doesn't need to be low hanging fruit. We are building a custom
supercomputer for this,
so we are interested how to speed up this workload.
In broad strokes, posits improve on IEEE float in reduced complexity,
and adherence to associative and distributive laws of arithmetic. The
latter is where
the reproducibility comes from.
The basic mechanism to produce reproducibility is by explicit management of
events. In a posit system, we can accumulate the unrounded output of the
and multipliers and if you say, 1/sqrt(x) is important we can accumulate
that unrounded too,
although we all know that is not quite a true statement when div is
Our initial thought was that this mechanism to accumulate large number of
and large values without error could be used advantageously for the force
Secondly, for forward/inverse transform calculations, this mechanism yields
a pure identity
matrix, and our FFT/iFFT take advantage of this and suddenly 16-bit posits
with 64-bit doubles.
If we need some education on how best to apply this to MM-style MDs, we are
to learn how best to apply this from GROMACS SMEs.
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the gromacs.org_gmx-developers