AMD Takes High-Performance Datacenter Computing to the Next Horizon

— Laun­ches World’s first 7nm High-per­for­mance GPU for Machi­ne Lear­ning and AI; Demons­tra­tes World’s First High-per­for­mance 7nm x86 CPU Powe­red by “Zen 2” Pro­ces­sor Core and Bre­akth­rough Chip­let Design —

— Ama­zon Web Ser­vices Announ­ces Imme­dia­te Avai­la­bi­li­ty of AMD EPYC pro­ces­sor-powe­red Ver­si­ons of Its Popu­lar Instan­ce Fami­lies —

SAN FRANCISCO

 

AMD (NASDAQ: AMD) today demons­tra­ted its total com­mit­ment to dat­a­cen­ter com­pu­ting inno­va­ti­on at its Next Hori­zon event in San Fran­cis­co by detail­ing its upco­m­ing 7nm com­pu­te and gra­phics pro­duct port­fo­lio desi­gned to extend the capa­bi­li­ties of the modern dat­a­cen­ter. During the event, AMD sha­red new spe­ci­fics on its upco­m­ing “Zen 2” pro­ces­sor core archi­tec­tu­re, detail­ed its revo­lu­tio­na­ry chip­let-based x86 CPU design, laun­ched the 7nm AMD Rade­on Instinct™ MI60 gra­phics acce­le­ra­tor and pro­vi­ded the first public demons­tra­ti­on of its next-genera­ti­on 7nm EPYC™ ser­ver pro­ces­sor code­na­med “Rome”. Ama­zon Web Ser­vices (AWS), the world’s most com­pre­hen­si­ve and broad­ly adop­ted cloud plat­form com­pa­ny, joi­ned AMD at the event to announ­ce the avai­la­bi­li­ty of three of its popu­lar instan­ce fami­lies on the Ama­zon Elastic Com­pu­te Cloud (EC2) powe­red by the AMD EPYC™ pro­ces­sor.

The mul­ti-year invest­ments we have made in our dat­a­cen­ter hard­ware and soft­ware road­maps are dri­ving gro­wing adop­ti­on of our CPUs and GPUs across cloud, enter­pri­se and HPC custo­mers,” said Dr. Lisa Su, pre­si­dent and CEO, AMD. “We are well posi­tio­ned to acce­le­ra­te our momen­tum as we intro­du­ce the industry’s broa­dest, most power­ful port­fo­lio of dat­a­cen­ter CPUs and GPUs fea­turing indus­try-lea­ding 7nm pro­cess tech­no­lo­gy over the com­ing quar­ters.”

AMD Compute Architecture Updates

AMD for the first time detail­ed its upco­m­ing “Zen 2” high-per­for­mance x86 CPU pro­ces­sor core that is the result of a revo­lu­tio­na­ry modu­lar design metho­do­lo­gy. This modu­lar sys­tem design uses an enhan­ced ver­si­on of AMD Infi­ni­ty Fabric inter­con­nect to link sepa­ra­te pie­ces of sili­con (“chip­lets”) wit­hin a sin­gle pro­ces­sor packa­ge. The mul­ti-chip pro­ces­sor uses 7nm pro­cess tech­no­lo­gy for the “Zen 2” CPU cores that bene­fit from the advan­ced pro­cess tech­no­lo­gy, while lever­aging a matu­re 14nm pro­cess tech­no­lo­gy for the input/output por­ti­on of the chip. The result is much hig­her per­for­mance – more CPU cores at the same power, and more cost-effec­tive manu­fac­tu­re than tra­di­tio­nal mono­li­thic chip designs.

Com­bi­ning this bre­akth­rough design metho­do­lo­gy with the bene­fits of TSMC’s lea­ding-edge 7nm pro­cess tech­no­lo­gy, “Zen 2” deli­vers signi­fi­cant per­for­mance, power con­sump­ti­on and den­si­ty genera­tio­nal impro­ve­ments that can help redu­ce dat­a­cen­ter ope­ra­ting costs, car­bon foot­print and coo­ling requi­re­ments. Other key genera­tio­nal advan­ces over the award-win­ning “Zen” core inclu­de:

  • An impro­ved exe­cu­ti­on pipe­line, fee­ding its com­pu­te engi­nes more effi­ci­ent­ly. 
  • Front-end advan­ces – impro­ved branch pre­dic­tor, bet­ter inst­ruc­tion pre-fet­ching, re-opti­mi­zed inst­ruc­tion cache and lar­ger op cache.
  • Floa­ting point enhan­ce­ments – dou­bled floa­ting point width to 256-bit and load/store band­width, increa­sed dispatch/retire band­width and main­tai­ned high through­put for all modes.
  • Advan­ced secu­ri­ty fea­tures – Hard­ware-enhan­ced Spec­t­re miti­ga­ti­ons, taking soft­ware migra­ti­on and har­de­ning it into the design, and increa­sed fle­xi­bi­li­ty of memo­ry encryp­ti­on.

Mul­ti­ple 7nm-based AMD pro­duc­ts are now in deve­lop­ment, inclu­ding next-genera­ti­on AMD EPYC CPUs and AMD Rade­on Instinct GPUs, both of which AMD detail­ed and demons­tra­ted at the event. Addi­tio­nal­ly, the com­pa­ny sha­red that its fol­low-on 7nm+-based “Zen 3” and “Zen 4” x86 core archi­tec­tures are on-track.

AMD EPYC Server CPU Updates

Rein­for­cing the gro­wing momen­tum achie­ved with its cur­rent-genera­ti­on AMD EPYC pro­ces­sors, Matt Gar­man, vice pre­si­dent of com­pu­te ser­vices at AWS joi­ned AMD on-sta­ge at the event to announ­ce the imme­dia­te avai­la­bi­li­ty of the first AMD EPYC pro­ces­sor-based instan­ces on Ama­zon Elastic Com­pu­te Cloud (EC2). Part of AWS’s popu­lar instan­ce fami­lies, the new AMD EPYC pro­ces­sor-powe­red offe­rings fea­ture indus­try-lea­ding core den­si­ty and memo­ry band­width. This results in excep­tio­nal per­for­mance-per-dol­lar for gene­ral pur­po­se and memo­ry-opti­mi­zed workloads, dri­ven by the core den­si­ty of AMD EPYC pro­ces­sors that offer M5a and T3a custo­mers a balan­ce of com­pu­te, memo­ry, and net­wor­king resour­ces for web and app­li­ca­ti­on ser­vers, backend ser­vers for enter­pri­se app­li­ca­ti­ons, and test/development envi­ron­ments with seam­less app­li­ca­ti­on migra­ti­on. For R5a custo­mers, the memo­ry band­width advan­ta­ge of AMD EPYC pro­ces­sors is ide­al for in-memo­ry pro­ces­sing, data mining, and dyna­mic data pro­ces­sing.

AMD also dis­c­lo­sed new details and deli­ve­r­ed per­for­mance pre­views of its next-genera­ti­on EPYC pro­ces­sors code­na­med “Rome”:

  • Pro­ces­sor enhan­ce­ments inclu­ding up to 64 “Zen 2” cores, increa­sed inst­ruc­tions-per-cycle1 and lea­dership com­pu­te, I/O and memo­ry band­width2.
  • Plat­form enhan­ce­ments inclu­ding the industry’s first PCIe 4.0-capable x86 ser­ver pro­ces­sor with dou­ble the band­width per chan­nel3 to dra­ma­ti­cal­ly impro­ve dat­a­cen­ter acce­le­ra­tor per­for­mance.
  • Dou­ble the com­pu­te per­for­mance per socket4 and four times the floa­ting point per­for­mance per socketcom­pa­red to cur­rent AMD EPYC pro­ces­sors.
  • Socket com­pa­ti­bi­li­ty with today’s AMD EPYC ser­ver plat­forms.

AMD demons­tra­ted the per­for­mance and plat­form advan­ta­ges of its next-genera­ti­on EPYC pro­ces­sor with two demos during the event:

  • A pre-pro­duc­tion sin­gle-socket next-genera­ti­on AMD EPYC pro­ces­sor out­per­forming a com­mer­ci­al­ly avail­ab­le top-of-the-line Intel dual pro­ces­sor Xeon ser­ver run­ning the com­pu­ta­tio­nal­ly-inten­si­ve, indus­try stan­dard “C-Ray” bench­mark6.
  • The industry’s first x86 PCIe 4.0-capable plat­form demo, fea­turing a Rade­on Instinct MI60 pro­ces­sor to acce­le­ra­te image reco­gni­ti­on.

Rome” is sam­pling with custo­mers now and is expec­ted to be the world’s first high-per­for­mance x86 7nm CPU.

AMD Datacenter Graphics Updates

AMD laun­ched the world’s first 7nm GPUs and the industry’s only hard­ware-vir­tua­li­zed GPUs – the AMD Rade­on Instinct MI60 and MI50 – which are sche­du­led to ship to custo­mers this quar­ter. The­se new gra­phics cards are based on the high-per­for­mance, fle­xi­ble “Vega” archi­tec­tu­re and are spe­ci­fi­cal­ly desi­gned for machi­ne lear­ning and arti­fi­ci­al intel­li­gence (AI), deli­vering hig­her levels of floa­ting-point per­for­mance7, grea­ter effi­ci­en­ci­esand new fea­tures for dat­a­cen­ter deploy­ments. A live demons­tra­ti­on during the event show­ed the flagship AMD Rade­on Instinct MI60 run­ning real-time trai­ning, infe­rence and image clas­si­fi­ca­ti­on.

In addi­ti­on to new hard­ware announ­ce­ments AMD also announ­ced ROCm 2.0, a new ver­si­on of its open soft­ware plat­form for acce­le­ra­ted com­pu­ting that inclu­des new math libra­ries, broa­der soft­ware frame­work sup­port, and opti­mi­zed deep lear­ning ope­ra­ti­ons. ROCm 2.0 has also been upstrea­med for Linux ker­nel dis­tri­bu­ti­ons, exten­ding ROCm access to mil­li­ons of Linux deve­lo­pers and users. Desi­gned for sca­le, ROCm allows custo­mers to deploy high-per­for­mance, ener­gy-effi­ci­ent hete­ro­ge­ne­ous com­pu­ting sys­tems in an open envi­ron­ment.

Pre­sen­ta­ti­ons from the event are avail­ab­le now on at www.amd.com/NextHorizon. A full replay will be avail­ab­le wit­hin 12 hours and will remain avail­ab­le for appro­xi­mate­ly one year.

Supporting Resources

About AMD

For more than 45 years AMD has dri­ven inno­va­ti­on in high-per­for­mance com­pu­ting, gra­phics and visua­li­za­ti­on tech­no­lo­gies ― the buil­ding blocks for gaming, immer­si­ve plat­forms and the dat­a­cen­ter. Hund­reds of mil­li­ons of con­su­mers, lea­ding For­tu­ne 500 busi­nes­ses and cut­ting-edge sci­en­ti­fic rese­arch faci­li­ties around the world rely on AMD tech­no­lo­gy dai­ly to impro­ve how they live, work and play. AMD employees around the world are focu­sed on buil­ding gre­at pro­duc­ts that push the bounda­ries of what is pos­si­ble. For more infor­ma­ti­on about how AMD is enab­ling today and inspi­ring tomor­row, visit the AMD (NASDAQ: AMDweb­siteblogFace­book and Twit­ter pages.

Cautionary Statement  

This press release con­tains for­ward-loo­king state­ments con­cer­ning Advan­ced Micro Devices, Inc. (AMD) inclu­ding the fea­tures, func­tio­n­a­li­ty, avai­la­bi­li­ty, timing, deploy­ment and expec­ta­ti­ons of AMD future pro­duc­ts and tech­no­lo­gies, inclu­ding “Zen 2,” “Zen 3,” “Zen 4,” “Rome,” AMD Rade­on Instinct™ MI60 and MI50 acce­le­ra­tors and the ROCm 2.0 open soft­ware plat­form; AMD dri­ving gro­wing adop­ti­on in its CPUs and GPUs across cloud, enter­pri­se and HPC custo­mers; AMD being well posi­tio­ned to acce­le­ra­te its momen­tum through the intro­duc­tion of AMD’s dat­a­cen­ter port­fo­lio in the com­ing quar­ters; and expec­ted bene­fits from EPYC™ based AWS R5, M5 and T3 instan­ces, which are made pur­suant to the Safe Har­bor pro­vi­si­ons of the Pri­va­te Secu­ri­ties Liti­ga­ti­on Reform Act of 1995. For­ward-loo­king state­ments are com­mon­ly iden­ti­fied by words such as “would,” “intends,” “belie­ves,” “expec­ts,” “may,” “will,” “should,” “seeks,” “intends,” “plans,” “pro for­ma,” “esti­ma­tes,” “anti­ci­pa­tes,” or the nega­ti­ve of the­se words and phra­ses, other varia­ti­ons of the­se words and phra­ses or com­pa­ra­ble ter­mi­no­lo­gy. Inves­tors are cau­tio­ned that the for­ward-loo­king state­ments in this docu­ment are based on cur­rent beliefs, assump­ti­ons and expec­ta­ti­ons, speak only as of the date of this docu­ment and invol­ve risks and uncer­tain­ties that could cau­se actu­al results to dif­fer mate­ri­al­ly from cur­rent expec­ta­ti­ons. Such state­ments are sub­ject to cer­tain known and unknown risks and uncer­tain­ties, many of which are dif­fi­cult to pre­dict and gene­ral­ly bey­ond AMD’s con­trol, that could cau­se actu­al results and other future events to dif­fer mate­ri­al­ly from tho­se expres­sed in, or implied or pro­jec­ted by, the for­ward-loo­king infor­ma­ti­on and state­ments. Mate­ri­al fac­tors that could cau­se actu­al results to dif­fer mate­ri­al­ly from cur­rent expec­ta­ti­ons inclu­de, wit­hout limi­ta­ti­on, the fol­lo­wing: Intel Corporation’s domi­nan­ce of the micropro­ces­sor mar­ket and its aggres­si­ve busi­ness prac­tices may limit AMD’s abi­li­ty to com­pe­te effec­tively; AMD has a wafer sup­ply agree­ment with GF with obli­ga­ti­ons to purcha­se all of its micropro­ces­sor and APU pro­duct requi­re­ments, and a cer­tain por­ti­on of its GPU pro­duct requi­re­ments, from GLOBALFOUNDRIES Inc. (GF) with limi­ted excep­ti­ons. If GF is not able to satis­fy AMD’s manu­fac­tu­ring requi­re­ments, its busi­ness could be adver­se­ly impac­ted; AMD reli­es on third par­ties to manu­fac­tu­re its pro­duc­ts, and if they are unab­le to do so on a time­ly basis in suf­fi­ci­ent quan­ti­ties and using com­pe­ti­ti­ve tech­no­lo­gies, AMD’s busi­ness could be mate­ri­al­ly adver­se­ly affec­ted; fail­u­re to achie­ve expec­ted manu­fac­tu­ring yields for AMD’s pro­duc­ts could nega­tively impact its finan­ci­al results; the suc­cess of AMD’s busi­ness is depen­dent upon its abi­li­ty to intro­du­ce pro­duc­ts on a time­ly basis with fea­tures and per­for­mance levels that pro­vi­de value to its custo­mers while sup­por­ting and coin­ci­ding with signi­fi­cant indus­try tran­si­ti­ons; if AMD can­not gene­ra­te suf­fi­ci­ent reve­nue and ope­ra­ting cash flow or obtain exter­nal finan­cing, it may face a cash short­fall and be unab­le to make all of its plan­ned invest­ments in rese­arch and deve­lop­ment or other stra­te­gic invest­ments; the loss of a signi­fi­cant custo­mer may have a mate­ri­al adver­se effect on AMD; AMD’s rece­ipt of reve­nue from its semi-custom SoC pro­duc­ts is depen­dent upon its tech­no­lo­gy being desi­gned into third-par­ty pro­duc­ts and the suc­cess of tho­se pro­duc­ts; AMD pro­duc­ts may be sub­ject to secu­ri­ty vul­nera­bi­li­ties that could have a mate­ri­al adver­se effect on AMD; data breaches and cyber-attacks could com­pro­mi­se AMD’s intel­lec­tu­al pro­per­ty or other sen­si­ti­ve infor­ma­ti­on, be cost­ly to reme­dia­te and cau­se signi­fi­cant dama­ge to its busi­ness and repu­ta­ti­on; AMD’s ope­ra­ting results are sub­ject to quar­ter­ly and sea­so­nal sales pat­terns; glo­bal eco­no­mic uncer­tain­ty may adver­se­ly impact AMD’s busi­ness and ope­ra­ting results; AMD may not be able to gene­ra­te suf­fi­ci­ent cash to ser­vice its debt obli­ga­ti­ons or meet its working capi­tal requi­re­ments; AMD has a lar­ge amount of indeb­ted­ness which could adver­se­ly affect its finan­ci­al posi­ti­on and pre­vent it from imple­men­ting its stra­te­gy or ful­fil­ling its con­trac­tu­al obli­ga­ti­ons; the agree­ments gover­ning AMD’s notes and the Secu­red Revol­ving Line of Credit impo­se restric­tions on AMD that may adver­se­ly affect its abi­li­ty to ope­ra­te its busi­ness; the mar­kets in which AMD’s pro­duc­ts are sold are high­ly com­pe­ti­ti­ve; AMD’s issu­an­ce to West Coast Hitech L.P. (WCH) of war­rants to purcha­se 75 mil­li­on sha­res of its com­mon stock, if and when exer­cis­ed, will dilu­te the ownership inte­rests of its exis­ting stock­hol­ders, and the con­ver­si­on of the 2.125% Con­ver­ti­ble Seni­or Notes due 2026 may dilu­te the ownership inte­rest of its exis­ting stock­hol­ders, or may other­wi­se depress the pri­ce of its com­mon stock; uncer­tain­ties invol­ving the orde­ring and ship­ment of AMD’s pro­duc­ts could mate­ri­al­ly adver­se­ly affect it; the demand for AMD’s pro­duc­ts depends in part on the mar­ket con­di­ti­ons in the indus­tries into which they are sold. Fluc­tua­ti­ons in demand for AMD’s pro­duc­ts or a mar­ket decli­ne in any of the­se indus­tries could have a mate­ri­al adver­se effect on its results of ope­ra­ti­ons; AMD’s abi­li­ty to design and intro­du­ce new pro­duc­ts in a time­ly man­ner is depen­dent upon third-par­ty intel­lec­tu­al pro­per­ty; AMD depends on third-par­ty com­pa­nies for the design, manu­fac­tu­re and sup­ply of mother­boards, soft­ware and other com­pu­ter plat­form com­pon­ents to sup­port its busi­ness; if AMD loses Micro­soft Corporation’s sup­port for its pro­duc­ts or other soft­ware ven­dors do not design and deve­lop soft­ware to run on AMD’s pro­duc­ts, its abi­li­ty to sell its pro­duc­ts could be mate­ri­al­ly adver­se­ly affec­ted; and AMD’s reli­an­ce on third-par­ty dis­tri­bu­tors and AIB part­ners sub­jec­ts it to cer­tain risks.  Inves­tors are urged to review in detail the risks and uncer­tain­ties in AMD’s Secu­ri­ties and Exchan­ge Com­mis­si­on filings, inclu­ding but not limi­ted to AMD’s Quar­ter­ly Report on Form 10-Q for the quar­ter ended Sep­tem­ber 29, 2018.