From: Jim Phillips (jim_at_ks.uiuc.edu)
Date: Fri Aug 05 2005 - 17:40:30 CDT
(Charm++ is the parallel runtime on which NAMD is based. -Jim)
4th Annual Workshop on Charm++ and its Applications 2005
Hosted by the Parallel Programming Laboratory
Department of Computer Science
University of Illinois at Urbana-Champaign
Charm++/AMPI and the frameworks built upon them have emerged as powerful
parallel programming systems in recent years. By allowing the programmer
to divide the computation into a large number of entities, which are mapped
to the available processors by an intelligent runtime system, Charm++ enables
a separation of concerns that leads to both improved productivity and higher
performance. Powerful abstractions are available in the FEM, POSE, MSA, and
Multiblock frameworks. Charm++ frameworks have enabled applications ranging
from molecular dynamics, cosmology, network simulation, materials science,
and rocket simulation.
The workshop will focus on showcasing the latest leading research in parallel
processing using Charm++ and its frameworks.
Topics will include advances in Parallel Simulation with POSE; Fault
Tolerance techniques; Communication Optimization libraries and frameworks;
Load Balancing strategies; as well as advances in AMPI/Charm++ primitives
and runtime system.
Authors and attendees are encouraged to share their experiences with the
systems built upon Charm++/AMPI as well as Charm++/AMPI itself. The workshop
will be a forum for developers and users of Charm++ to interact closely.
The workshop is also aimed at taking stock of the funded collaborative
interdisciplinary research projects involving the Charm++ system.
There will be several tutorials demonstrating the different features of
Charm++. The tutorials will be presented by the Charm++ developers. Attendees
are required to register for the tutorials. Following is the tentative list
· Charm++ and virtualization tutorial
· How to write applications using Adaptive MPI
· Using POSE for Parallel Discrete Event Simulation
· Running your application on BigSim
· Projections: Performance analysis
· FEM framework
· How to install and use the Faucets adaptive scheduler
Call For Papers:
Authors are invited to submit short manuscripts that demonstrate research
and development in all areas of parallel processing using Charm++ and AMPI.
Submissions should be in the form of extended abstracts up to 3 pages
formatted in standard IEEE format. The papers will not be published and the
authors are encouraged to publish them in other conferences. Some of the
topics of interest are:
· Parallel and Distributed Applications in science and engineering
· Parallel Discrete Event Simulation techniques and applications
· Parallel libraries and frameworks
· Extensions and new features in Charm++
· Massively parallel processing on future machines
· Novel parallel programming models
Please send submissions to Eric Bohm (ebohm_at_uiuc.edu) by the deadline below.
· Manuscripts due: September 2, 2005
· Notification of acceptance: September 9, 2005
· Workshop: October 17-19, 2005
This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:39:48 CST