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.
An highly efficient MGPT implementation for LAMMPS, with strong scaling
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):Tomas Oppelstrup, Alexander F. Stukowski, Jaime Marian
ROOM:WSCC North Galleria 2nd/3rd Floors
ABSTRACT: The MGPT potential has been implemented as a package for the general molecular dynamics code LAMMPS. We use an improved force calculation scheme which reduces the communication layer thickness by a factor of two, and makes the code more load balanced. This results in unprecedented strong scaling, and speedup continuing beyond 1/8 atom/core. In the strong scaling limit, our code is 10 times faster than the Gordon-Bell Prize winning BlueGene/L implementation of MGPT (2).
In addition, we have optimized the small matrix linear algebra, which accounts for most of the computaitonal work in MGPT. We generate kernels with different blocking parameters, and SIMD intrinsics for vectorization on x86 and BlueGene cpus, and select the fastest routines. On recent Intel cpus, the kernels useful flops/s reach 77% to theoretical peak performance.
The poster will illustrate our force calculation scheme, our approach to linear algebra optimization and show scaling and performance results.
Chair/Author Details:
Bernd Mohr (Chair) - Juelich Supercomputing Centre
Tomas Oppelstrup - Lawrence Livermore National Laboratory
Alexander F. Stukowski - Lawrence Livermore National Laboratory
Jaime Marian - Lawrence Livermore National Laboratory