SCHEDULE: NOV 16-22, 2013
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.
Optimizing MPI Implementation on Massively Parallel Many-Core Architectures
SESSION: Doctoral Showcase - Early Research Showcase
EVENT TYPE: Doctoral Showcase - Early Research Showcases
TIME: 3:30PM - 5:00PM
SESSION CHAIR: Wojtek James Goscinski
Presenter(s):Min Si
ROOM:601/603
ABSTRACT:
Many-core architectures, such as the Intel Xeon Phi, provide user
application dozens of cores and hundreds of hardware threads. To
efficiently utilize such architectures, application programmers are
increasingly looking at hybrid programming models, where multiple
threads often work with MPI, which is responsible for data
communication. My doctoral research focuses on exploiting the
capabilities of such massively threaded architectures in MPI. It can
be broadly divided into three related pieces: internal multithreading
in MPI, fine-grained consistency management in MPI, and MPI tasklets.
We have so far focused on the first topic, enabling MPI to share idle
threads from user applications. We parallelized the MPI functionality
used in stencil computation routines as a case study. Preliminary
results show that the parallel version provides up to a 40-fold
speedup in MPI performance when parallelized across the entire Intel
Xeon Phi board.
Chair/Presenter Details:
Wojtek James Goscinski (Chair) - Monash University
Min Si - University of Tokyo
Click here to download .ics calendar file