From: The lattice QCD executive committee
This is a draft of the consensus statement as agreed upon by the members of the lattice QCD executive committee (R. Sugar (chair), N. Christ, M. Creutz, P. Mackenzie, J. Negele, C. Rebbi, and S. Sharpe). Please send any comments to Bob Sugar (firstname.lastname@example.org).
Brookhaven National Laboratory hosted a workshop (``Workshop on a New Computing Venue for Lattice Gauge Theory Computing'', October 12-14, 2000) to acquaint US lattice gauge theorists with the potential advantages of the QCDOC computing architecture as a possible community resource in a BNL-located venue. The aim was to obtain feedback from the community, as well as from the wider US HENP community, on the ways in which a new QCDOC facility at BNL could best meet their needs, and to engage the community's support in seeking approval and funding of this facility by DOE.
The following consensus statement was agreed upon by attendees at the meeting. This statement refers to the White Paper written in 1999 by the Lattice QCD Executive Committee, in which a five-year plan for computing facilities for LQCD was enunciated.
Lattice simulations of QCD can potentially answer a variety of physics questions crucial to the interpretation of experimental results from the high energy and nuclear physics programs. These include results for electroweak matrix elements involving B, D, and K mesons, the mapping of the phase diagram for QCD at finite temperature and the calculation of the associated equation of state, and the study of the structure and interactions of hadrons. These three areas--weak matrix elements, finite temperature studies, and hadron physics--are closely associated with the physics at the DOE's facilities at FNAL, BNL and JLAB, respectively.
A computational resource enabling multiple calculations at the Teraflops-year scale will allow reliable results for several of the above mentioned quantities. It should be recognized, however, that there is a range of difficulty of calculations, and that to fully exploit the potential of LQCD calculations even more powerful computers will be needed in future years.
It is crucial for the US to obtain resources of this magnitude as soon as possible in order to remain competitive with European and Japanese lattice gauge theorists.
In order to attain this goal, we support the overall process outlined in the White Paper. In particular, we endorse the following program, tuned to make use of the opportunity presented by the SDAC initiative. We stress that an important aspect of this plan is to study and exploit different computer technologies so as to make the most cost-effective choices for simulating LQCD.