SC is the International Conference for
High Performance Computing, Networking,
Storage and Analysis



SCHEDULE: NOV 12-18, 2011

When viewing the Technical Program schedule, on the far righthand side is a column labeled "PLANNER." Use this planner to build your own schedule. Once you select an event and want to add it to your personal schedule, just click on the calendar icon of your choice (outlook calendar, ical calendar or google calendar) and that event will be stored there. As you select events in this manner, you will have your own schedule to guide you through the week.

You can also create your personal schedule on the SC11 app (Boopsie) on your smartphone. Simply select a session you want to attend and "add" it to your plan. Continue in this manner until you have created your own personal schedule. All your events will appear under "My Event Planner" on your smartphone.

3D Tixels: A Highly Efficient Algorithm for GPU/CPU-Acceleration of Molecular Dynamics on Heterogeneous Parallel Architectures

SESSION: Research Poster Reception

EVENT TYPE: ACM Student Research Competition Poster, Poster, Electronic Poster

TIME: 5:15PM - 7:00PM

SESSION CHAIR: Bernd Mohr

AUTHOR(S):Szilárd Páll, Berk Hess, Erik Lindahl

ROOM:WSCC North Galleria 2nd/3rd Floors

ABSTRACT:
Several GPU-based algorithms have been developed to accelerate biomolecular simulations, but although they provide benefits over single-core implementations, they have not been able to beat state-of-the art SIMD CPU implementations (e.g. GROMACS) running on modern hardware with 12-24 cores, not to mention achieve efficient scaling. Here, we present a new heterogeneous acceleration approach that utilizes all CPU/GPU resources efficiently. A novel fixed-particle-number sub-cell algorithm makes it possible to combine neutral-territory domain decomposition in GROMACS with CUDA nonbonded kernels that achieve 60% work-efficiency, 1.5IPC and 95% cache utilization. The algorithm is intrinsically future-proof since cell sizes can be adapted to future hardware. Lattice summation is simultaneously handled with an OpenMP-based CPU implementation that balances the load between CPU/GPU domains. We achieve threefold speedup compared to GROMACS running on all cores of a workstation, which to the best of our knowledge makes this the fastest GPU molecular dynamics implementation presented to date.

Chair/Author Details:

Bernd Mohr (Chair) - Juelich Supercomputing Centre

Szilárd Páll - KTH Royal Institute of Technology

Berk Hess - KTH Royal Institute of Technology

Erik Lindahl - KTH Royal Institute of Technology

Add to iCal  Click here to download .ics calendar file

Add to Outlook  Click here to download .vcs calendar file

Add to Google Calendarss  Click here to add event to your Google Calendar

   Sponsors    ACM    IEEE