Last update: Apr 1, 2000
Speaker: Donald HolmgrenLarge scale QCD Monte Carlo calculations have typically been performed on either commercial supercomputers or specially built massively parallel computers such as Fermilab's ACPMAPS.
Commodity computer systems offer impressive floating point performance at a cost which rapidly outstrips the performance to price ratio of commercial supercomputers. As high performance networking components become available at comparable cost, it has become reasonable to assemble a massively parallel supercomputer from commodity parts.
We describe the work and progress to date of a collaboration working on this problem. It is made up of high energy theorists and computing professionals from the Fermilab Theoretical Physics Group, the Fermilab Computing Division, the MILC Collaboration and the Cornell Particle Theory Group.
|| | Home | Bulletins | Committees | Scientific Program | Docs by topics | Social Event | Conference Location | Secretariat ||