LANL Open Supercomputing User Group           

An attempted grass-roots effort to form a users group. Failed due to LANL not having any open supercomputing at the time (maybe we can start again now?) and, even more sadly, due to lack of support.

Los Alamos has a long and distinguished history in advancing the state-of-the-art in computing hardware, software, and numerical algorithms, and in developing applications for solving challenging scientific and engineering problems. To help maintain this position in the supercomputing vanguard, and to foster an environment that promotes these activities at the Laboratory, we are in the process of forming a LANL open supercomputing user group.

The start-up committee decided to first have a ``virtual'' meeting on the web involving everyone who responded to the initial call for participation in the user group. This meeting is now closed. To go through the responses please go to the First Virtual Meeting Responses page.

The first organizational meeting of the User Group will be held in the Center for Nonlinear Studies (CNLS) [TA-3, SM1690] Conference Room at 1.00, on Wednesday, Feb. 17. All are encouraged to attend.

Membership in the user group is voluntary and open to all LANL and non-LANL users of the Lab's open supercomputing facilities, such as machines at the Advanced Computing Laboratory (ACL) and Avalon, the T-Division/CNLS DEC Alpha cluster. In addition, LANL staff members who use supercomputing facilities elsewhere (e.g., NERSC, NCAR, NCSA, --) are also encouraged to join.

Provide communication channel between users to prevent and reduce fragmentation of effort, to help inexperienced users, and to exchange useful information on all aspects of supercomputing. Provide easily accessible long-term ``institutional memory'' via web archive on these issues and the ones listed below.

Provide communication channel between users and Lab management with regard to open supercomputing priorities in general (both science and computing related), institutional thrusts such as the Institutional High Performance Computing Initiative (IHPCI), and long-term open supercomputing policy. Function as a forum (via task forces and working groups) to provide information needed by management on topics such as future hardware, peer review of open supercomputing proposals, and best mix of local vs. center-based computing options.

Provide communication channel between users and managers of open computing resources with regard to operational issues such as allocations, queue structures, and accounting. Provide forum for rationalization of procedures given various constraints (e.g., the fact that big projects need to get work done on tight schedules).

To join the user group, go to the open supercomputing user group membership form. There is sufficient space for questions, comments and suggestions (please do not send email to members of the start-up committee).

The initial organization of the user group is being conducted by a start-up committee consisting of Susan Coghlan (ACL), Salman Habib (T-8), John Pearson (XCM), Robert Ryne (LANSCE-1), Mike Warren (T-6), and Judy Winterkamp (EES-8). All issues regarding the final organization of the user group will be decided after membership becomes sufficient to provide a representative sample of open supercomputing users.

Salman Habib / LANL / habib@lanl.gov / revised December 98