- Home
- Register
- Attend
- Conference Program
- SC15 Schedule
- Technical Program
- Awards
- Students@SC
- Research with SCinet
- HPC Impact Showcase
- HPC Matters Plenary
- Keynote Address
- Support SC
- SC15 Archive
- Exhibits
- Media
- SCinet
- HPC Matters
SCHEDULE: NOV 15-20, 2015
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.
Improving Concurrency and Asynchrony in Multithreaded MPI Applications using Software Offloading
SESSION: MPI/Communication
EVENT TYPE: Papers
EVENT TAG(S): Power, System Software, Clouds and Distributed Computing, Resiliency
TIME: 4:00PM - 4:30PM
SESSION CHAIR(S): Yong Chen
AUTHOR(S):Karthikeyan Vaidyanathan, Dhiraj D. Kalamkar, Kiran Pamnany, Jeff R. Hammond, Pavan Balaji, Dipankar Das, Jongsoo Park, Balint Joo
ROOM:18CD
ABSTRACT:
We present a new approach for multithreaded communication
and asynchronous progress in MPI applications, wherein we offload
communication processing to a dedicated thread. The central
premise is that given the rapidly increasing core counts on modern
systems, the improvements in MPI performance arising from
dedicating a thread to drive communication outweigh the small
loss of resources for application computation, particularly when
overlap of communication and computation can be exploited. Our
approach allows application threads to make MPI calls concurrently,
enqueuing these as communication tasks to be processed
by a dedicated communication thread. This not only guarantees
progress for such communication operations, but also reduces load
imbalance. Our implementation
additionally significantly reduces the overhead of mutual
exclusion seen in existing implementations for applications using
MPI THREAD MULTIPLE. Our technique requires no modification
to the application, and we demonstrate significant performance
improvement (up to 2X) for QCD, FFT and deep learning CNN.
Chair/Author Details:
Yong Chen (Chair) - Texas Tech University|
Karthikeyan Vaidyanathan - Intel Corporation
Dhiraj D. Kalamkar - Intel Corporation
Kiran Pamnany - Intel Corporation
Jeff R. Hammond - Intel Corporation
Pavan Balaji - Argonne National Laboratory
Dipankar Das - Intel Corporation
Jongsoo Park - Intel Corporation
Balint Joo - Thomas Jefferson National Accelerator Facility
Click here to download .ics calendar file
Click here to download .vcs calendar file
Click here to add event to your Google Calendar