AMD Awarded $32 Million for ‘Extreme Scale’ High-Performance Computing Research Focused on HSA, APUs and Memory

AMD to Expand Rese­arch on Exas­ca­le Com­pu­ters for the Depart­ment of Energy

SUNNYVALE, CA — (Mar­ket­wired) — 11/14/14 – AMD (NYSE: AMD) today announ­ced that for the third straight year it was award­ed rese­arch grants for deve­lo­p­ment of cri­ti­cal tech­no­lo­gies nee­ded for extre­me-sca­le com­pu­ting in con­junc­tion with pro­jects asso­cia­ted with the U.S. Depart­ment of Ener­gy (DOE) Extre­me-Sca­le Com­pu­ting Rese­arch and Deve­lo­p­ment Pro­gram, known as “Fast­For­ward 2.” The two DOE awards, tota­ling more than $32 mil­li­on, will fund rese­arch focu­sed on exas­ca­le appli­ca­ti­ons for AMD Acce­le­ra­ted Pro­ces­sing Units (APUs) based on the open-stan­dard Hete­ro­ge­neous Sys­tem Archi­tec­tu­re (HSA), as well as future memo­ry sys­tems to power a gene­ra­ti­on of exas­ca­le super­com­pu­ters capa­ble of deli­ve­ring 30–60 times more per­for­mance than today’s fas­test supercomputers.

Fast­For­ward 2 is a joint­ly-fun­ded col­la­bo­ra­ti­on bet­ween DOE Office of Sci­ence and the Natio­nal Nuclear Secu­ri­ty Admi­nis­tra­ti­on (NNSA) focu­sed on initia­ting part­ner­ships with mul­ti­ple com­pa­nies to acce­le­ra­te the rese­arch and deve­lo­p­ment of cri­ti­cal tech­no­lo­gies nee­ded to enable exas­ca­le com­pu­ting. Spe­ci­fi­cal­ly, the Fast­For­ward 2 rese­arch grants award­ed to AMD are for:

  • Node Archi­tec­tu­re: AMD will con­duct rese­arch for an inte­gra­ted exas­ca­le node archi­tec­tu­re based on its HSA-enab­led APUs. Par­ti­cu­lar are­as of rese­arch include power effi­ci­en­cy, relia­bi­li­ty, pro­gramma­bi­li­ty, com­po­nent and net­work inter­face inte­gra­ti­on, APU micro­ar­chi­tec­tu­re, advan­ced memo­ry archi­tec­tures and effi­ci­ent data movement.
  • Memo­ry Tech­no­lo­gy: AMD will col­la­bo­ra­te with the DOE and others to help defi­ne a new stan­dard for memo­ry inter­faces that meets the needs of future-gene­ra­ti­on memo­ry devices, inclu­ding non-vola­ti­le memo­ry and pro­ces­sing-in-memo­ry (PIM) architectures.

We’­re hono­red to be sel­ec­ted once again by DOE to aid rese­arch efforts for exas­ca­le-com­pu­ting,” said Alan Lee, AMD’s cor­po­ra­te vice pre­si­dent for Rese­arch and Advan­ced Deve­lo­p­ment. “Our cus­to­mers have long loo­ked to AMD to deve­lop inno­va­ti­ve com­pu­te, gra­phics, and memo­ry tech­no­lo­gies and pro­ducts that sol­ve real-world pro­blems. This rese­arch is focu­sed on ener­gy-effi­ci­ent node archi­tec­tures and memo­ry sys­tems to impro­ve the capa­bi­li­ty of the world’s fas­test supercomputers.”

The DOE stra­te­gic plan seeks to address the nation’s most pres­sing sci­en­ti­fic chal­lenges by advan­cing simu­la­ti­on-based sci­en­ti­fic dis­co­very made pos­si­ble by the world’s hig­hest per­forming exas­ca­le super­com­pu­ters. The tech­no­lo­gies deve­lo­ped as part of this initia­ti­ve could posi­tively impact low-power embedded, cloud/datacenter and mid-ran­ge high-per­for­mance com­pu­ting appli­ca­ti­ons for diver­se fields ran­ging from medi­cal sci­ence to astro­phy­sics and cli­ma­te mode­ling. Fast­For­ward 2 invest­ments will result in an increase in DOE’s abili­ty to levera­ge com­mer­cial deve­lo­p­ments for future systems.

AMD Opte­ron™ com­pu­te pro­ces­sors and AMD Fire­Pro™ gra­phics pro­ces­sors are used today in many of the world’s lea­ding super­com­pu­ters.

About AMD
AMD (NYSE: AMD) designs and inte­gra­tes tech­no­lo­gy that powers mil­li­ons of intel­li­gent devices, inclu­ding per­so­nal com­pu­ters, tablets, game con­so­les and cloud ser­vers that defi­ne the new era of sur­round com­pu­ting. AMD solu­ti­ons enable peo­p­le ever­y­whe­re to rea­li­ze the full poten­ti­al of their favo­ri­te devices and appli­ca­ti­ons to push the boun­da­ries of what is pos­si­ble. For more infor­ma­ti­on, visit www.amd.com.

AMD, the AMD Arrow logo, AMD Opte­ron, AMD Fire­Pro and com­bi­na­ti­ons the­reof, are trade­marks of Advan­ced Micro Devices, Inc. Other names are for infor­ma­tio­nal pur­po­ses only and may be trade­marks of their respec­ti­ve owners.