SC12 Home > SC12 Schedule > SC12 Presentation - A Framework for Low-Communication 1-D FFT

SCHEDULE: NOV 10-16, 2012

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.

A Framework for Low-Communication 1-D FFT

SESSION: Fast Algorithms

EVENT TYPE: Papers, Best Paper Finalists

TIME: 10:30AM - 11:00AM

SESSION CHAIR: Torsten Hoefler

AUTHOR(S):Ping Tak Peter Tang, Jongsoo Park, Daehyun Kim, Vladimir Petrov

ROOM:255-BC

ABSTRACT:
In high-performance computing on distributed-memory systems, communication often represents a significant part of the overall execution time. The relative cost of communication will certainly continue to rise as compute density growth follows the current technology and industry trends. Design of lower-communication alternatives to fundamental computational algorithms has become an important field of research. For distributed 1-D FFT, communication cost has hitherto remained high as all industry-standard implementations perform three all-to-all internode data exchanges (also called global transpose). These communication steps indeed dominate execution time. In this paper, we present a mathematical framework from which many single-all-to-all and easy-to-implement 1-D FFT algorithms can be derived. For large-scale problems, our implementation can be twice as fast as leading FFT libraries on state-of-the-art computer clusters. Moreover, our framework allows tradeoff between accuracy and performance, further boosting performance if reduced accuracy is acceptable.

Chair/Author Details:

Torsten Hoefler (Chair) - ETH Zurich

Ping Tak Peter Tang - Intel Corporation

Jongsoo Park - Intel Corporation

Daehyun Kim - Intel Corporation

Vladimir Petrov - Intel Corporation

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

A Framework for Low-Communication 1-D FFT

SESSION: Fast Algorithms

EVENT TYPE: , Best Paper Finalists

TIME: 10:30AM - 11:00AM

SESSION CHAIR: Torsten Hoefler

AUTHOR(S):Ping Tak Peter Tang, Jongsoo Park, Daehyun Kim, Vladimir Petrov

ROOM:255-BC

ABSTRACT:
In high-performance computing on distributed-memory systems, communication often represents a significant part of the overall execution time. The relative cost of communication will certainly continue to rise as compute density growth follows the current technology and industry trends. Design of lower-communication alternatives to fundamental computational algorithms has become an important field of research. For distributed 1-D FFT, communication cost has hitherto remained high as all industry-standard implementations perform three all-to-all internode data exchanges (also called global transpose). These communication steps indeed dominate execution time. In this paper, we present a mathematical framework from which many single-all-to-all and easy-to-implement 1-D FFT algorithms can be derived. For large-scale problems, our implementation can be twice as fast as leading FFT libraries on state-of-the-art computer clusters. Moreover, our framework allows tradeoff between accuracy and performance, further boosting performance if reduced accuracy is acceptable.

Chair/Author Details:

Torsten Hoefler (Chair) - ETH Zurich

Ping Tak Peter Tang - Intel Corporation

Jongsoo Park - Intel Corporation

Daehyun Kim - Intel Corporation

Vladimir Petrov - Intel Corporation

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