"The cloud" - on-demand distributed computing power

Andrew Walkingshaw — Sat, 10 May 2008 16:25:00 -0700

Description

Simulation scientists are mostly limited by both the number, and the speed, of the computers available to them. Really large simulations need really serious computer resources, but simulations like that are pretty rare; so the resources for them have been concentrated in regional grids or national centres like the UK's HPCx.

Economically this makes a lot of sense, but there's a lot of overhead; for instance, compute time has to be bid for long in advance of when it might actually be used. In that light, commodity on-demand computing services like Amazon's EC2 begin to look promising as an alternative; they have even greater economies of scale than the national infrastructure services, can provide a scientist with more CPU power at essentially no notice, and often provide more flexibility in choice of operating system and software than a centrally-provided system can. At the moment, they don't scale to the massively parallel calculations that the national supercomputers specialize in, but sooner or later they'll be competitive even for those cases.

Computer & Information Science

Source

http://www.grid-support.ac.uk/content/view/239/157/
http://www.hpcx.ac.uk/
http://aws.amazon.com/ec2

Really, really big atomistic simulations

Andrew Walkingshaw — Fri, 02 May 2008 11:43:26 -0700

Description

Computers - or, at least, rooms full of computers - are really scarily fast nowadays, and faced with a fast computer, chemical physicists start running atomistic simulations on it.

DL_POLY_3 has set the record for largest system simulated by a general MD package. In March 2006 it ran a simulation of a 300 million atom system of NaCl with full Ewald sumation of the coulombic interactions on 1024 CPUs with 2GB/CPU RAM of the HPCx cluster.

To put this into perspective, that's a cube of atoms around 1000 angstroms across - only a factor of two smaller than the resolution of an optical microscope. We're getting to the point where simulation of mesoscale systems, and therefore seriously large-scale materials dynamics, is genuinely possible; and it's not too far from where a lot of biological systems suddenly come into play. These codes, roughly speaking, take an amount of time (per MD timestep) proportional to the number of atoms they're trying to simulate, so a factor of eight speedup in computer performance will translate to a box 0.2 microns across - fully atomistic simulations of crystals you can see.

Couple that with free-energy surface probes like metadynamics or ART Nouveau, and the prediction of properties - and ab-initio design - of a whole range of technological/nanomaterials beckons.

Future of chemistry

Source

http://www.eminerals.org/about/codes/dlpoly.html
http://people.sissa.it/~laio/metadynamics.htm
http://www.mrs.org/s_mrs/sec_subscribe.asp?CID=2590&DID=109618&action=detail

molecular dynamics

"The cloud" - on-demand distributed computing power

Description

Source

Really, really big atomistic simulations

Description

Source