Back to Results
First PageMeta Content
Computational chemistry / Constraint algorithm / Co-array Fortran / Science / Application programming interfaces / Message Passing Interface / Computational physics / Computational science / Physics


Advanced Communication Techniques for Gyrokinetic Fusion Applications on Ultra-Scale Platforms Robert Preissl John Shalf Alice Koniges Lawrence Berkeley National Laboratory
Add to Reading List

Document Date: 2011-10-10 09:37:27

Open Document

File Size: 2,74 MB

Share Result on Facebook

City

Supercomputing / Budapest / New York / Lugano / Washington / DC / London / /

Company

IBM / IEEE Computer / John Shalf Alice Koniges Lawrence Berkeley National Laboratory / IEEE Computer Society Press / AMD / Cray User Group / Bill Long Stephane Ethier Princeton Plasma Physics Laboratory / /

Country

Italy / United States / United Kingdom / /

Currency

AMD / USD / /

/

Facility

National Energy Research Scientific Computing Center / Ultra-Scale Platforms Robert Preissl John Shalf Alice Koniges Lawrence Berkeley National Laboratory / University of California at Berkeley / GASNet library / /

IndustryTerm

32K processors / lower startup / quicksort algorithm / exascale computing / presented algorithm / ultra-scale systems / parallel computing languages / torus network / communication technology / fusion energy / particle shift algorithm / 131K processors / 3D torus network / magnetic fusion devices / communication algorithms / 4K processors / communication intense applications / innovations in algorithms / Particle shift algorithms / energy / magnetic confinement devices / 23K processors / lower software overhead / 16K processors / message passing algorithms / shift algorithm / stage shift algorithm / communications approaches / 32K processing cores / shift algorithms / particle movement algorithms / Parallel numerical algorithms / tokamak toroidal devices / /

MarketIndex

NAS FT / CAF / /

OperatingSystem

Linux / /

Organization

University of California / U.S. Department of Energy / office of Science / ASIC / National Energy Research Scientific Computing Center / Department of Defense / IEEE Computer Society / Cray Center of Excellence / Ultra-Scale Platforms Robert Preissl John Shalf Alice Koniges Lawrence Berkeley National Laboratory / /

Person

Nicholas Wright / Tarek A. ElGhazawi / Yuri Dotsenko / J. N. Leboeuf / V / Mohamed M. Zahran / Guohua Jin / Robert W. Numrich / Richard Barrett / Dan Bonachea / Jason Eckhardt / Francois Cantonnet / Katherine Yelick / B. A. Carreras / Cristian Coarfa / Kim Kieun / Laksono Adhianto / Tarek El-Ghazawi / William N. Scherer III / Christian Bell / Reid / Rajesh Nishtala / Terry Clark / Yiyi Yao / Jason Duell / Ion Temperature Gradient-Driven / Scott L. Ridgway / Piotr Bala / Nathan Wichmann / /

/

Position

translator / Director / model for the communication / /

Product

GTS / Franklin / Linux / XE6 system / Cray XE6 / /

ProgrammingLanguage

Fortran 2008 / Fortran 90 / Fortran / DC / /

ProvinceOrState

New York / California / /

Technology

revised CAF algorithm / previous MPI algorithm / CAF-lock algorithm / Opteron processors / 4K processors / Coarray algorithm / existing MPI algorithm / quicksort algorithm / Linux / using 32K processors / CAF algorithm / message passing algorithms / fluid dynamics / stage shift algorithm / MPI-ss shift algorithm / communication technology / 16K processors / CAF algorithms / shared memory / MPI communication algorithms / 32K processors / light-weight one-sided messaging algorithms / communication algorithms / PGAS algorithm / using 131K processors / Particle shift algorithms / ASIC / Coarray particle shift algorithms / CAFlock algorithm / shift algorithm / 32768 processors / using 32 / 768 processors / two-sided particle shift algorithm / particle movement algorithms / 23K processors / MPI-2 algorithm / MPI shift algorithms / 131K processors / CAF particle shift algorithms / MPI shift algorithm / PIC algorithm / performing Coarray algorithm / Gemini router / MPI algorithms / Opteron processor / simulation / MPI-ms algorithm / MPI-ss algorithm / Parallel numerical algorithms / four algorithms / using 16K processors / Parallel Processing / presented algorithms / /

SocialTag