— AMD laun­ches AMD Instinct™ MI200 series acce­le­ra­tors, pre­views 3rd Gen AMD EPYC™ pro­ces­sors with AMD 3D V‑Cache, and pro­vi­des new details on expan­ded set of next-gene­ra­ti­on EPYC™ pro­ces­sors powered by “Zen 4” and “Zen 4c” CPU cores—

— Meta choo­ses EPYC™ CPUs for its data center —

SANTA CLARA, Calif., Nov. 08, 2021 (GLOBE NEWSWIRE) — AMD (NASDAQ: AMD) held the vir­tu­al Acce­le­ra­ted Data Cen­ter Pre­mie­re, laun­ching the new AMD Instinct™ MI200 series acce­le­ra­tors, the world’s fas­test acce­le­ra­tor for high per­for­mance com­pu­ting (HPC) and arti­fi­ci­al intel­li­gence (AI) workload­si, and pro­vi­ded a pre­view of the inno­va­ti­ve 3rd Gen AMD EPYC™ pro­ces­sors with AMD 3D V‑Cache. AMD also reve­a­led new infor­ma­ti­on about its next gene­ra­ti­on “Zen 4” pro­ces­sor core and announ­ced the new “Zen 4c” pro­ces­sor core, both of which will power future AMD ser­ver pro­ces­sors and are desi­gned to extend the company’s lea­der­ship pro­ducts for the data center.

“We are in a high-per­for­mance com­pu­ting mega­cy­cle that is dri­ving demand for more com­pu­te to power the ser­vices and devices that impact every aspect of our dai­ly lives,” said Dr. Lisa Su, pre­si­dent and CEO, AMD. “We are buil­ding signi­fi­cant momen­tum in the data cen­ter with our lea­der­ship pro­duct port­fo­lio, inclu­ding Meta’s adop­ti­on of AMD EPYC to power their infra­struc­tu­re and the buil­dout of Fron­tier, the first U.S. exas­ca­le super­com­pu­ter which will be powered by EPYC and AMD Instinct pro­ces­sors. In addi­ti­on, today we announ­ced a breadth of new pro­ducts that build on that momen­tum in next-gene­ra­ti­on EPYC pro­ces­sors with new inno­va­tions in design, lea­der­ship, 3D pack­a­ging tech­no­lo­gy, and 5 nm high-per­for­mance manu­fac­tu­ring to fur­ther extend our lea­der­ship for cloud, enter­pri­se and HPC customers.”

Meta Adopts EPYC CPUs [03:09 – 05:29]
AMD announ­ced Meta is the latest major hypers­ca­le cloud com­pa­ny that has adopted AMD EPYC CPUs to power its data cen­ters. AMD and Meta work­ed tog­e­ther to defi­ne an open, cloud-sca­le, sin­gle-socket ser­ver desi­gned for per­for­mance and power effi­ci­en­cy, based on the 3rd Gen EPYC pro­ces­sor. Fur­ther details will be dis­cus­sed at the Open Com­pu­te Glo­bal Sum­mit later this week. 

Advan­ced Pack­a­ging Dri­ving Data Cen­ter Per­for­mance [05:35 – 18:00]
AMD pre­view­ed the use of inno­va­ti­ve 3D chip­let pack­a­ging tech­no­lo­gy in the data cen­ter with the first ser­ver CPU using high-per­for­mance 3D die stack­ing. The 3rd Gen AMD EPYC pro­ces­sors with AMD 3D V‑Cache, code­na­med “Milan‑X,” repres­ents an inno­va­ti­ve step for­ward in CPU design and pack­a­ging, and will offer a 50% avera­ge per­for­mance uplift across tar­ge­ted tech­ni­cal com­pu­ting workloadsii.

3rd Gen EPYC with AMD 3D V‑Cache will offer the same capa­bi­li­ties and fea­tures as the 3rd Gen EPYC pro­ces­sors and they will be drop-in com­pa­ti­ble with a BIOS upgrade, deli­ve­ring easy adop­ti­on and per­for­mance enhancements.
Micro­soft Azu­re HPC vir­tu­al machi­nes fea­turing 3rd Gen EPYC with AMD 3D V‑Cache are available today in Pri­va­te Pre­view, with broad roll­out in the coming weeks. More infor­ma­ti­on on per­for­mance and avai­la­bi­li­ty is available here.
3rd Gen EPYC CPUs with AMD 3D V‑Cache will launch in Q1 2022. Part­ners inclu­ding Cis­co, Dell Tech­no­lo­gies, Leno­vo, HPE and Super­mi­cro are plan­ning to offer ser­ver solu­ti­ons with the­se processors.

Deli­ve­ring Exas­ca­le Class Per­for­mance for Acce­le­ra­ted Com­pu­ting [18:02 – 31:50]
AMD laun­ched the AMD Instinct MI200 series acce­le­ra­tors. Based on the AMD CDNA™ 2 archi­tec­tu­re, the MI200 series acce­le­ra­tors are the most advan­ced acce­le­ra­tors in the worl­diii and pro­vi­de up to 4.9x hig­her peak per­for­mance for HPC workload­siv and 1.2X hig­her peak flops of mixed pre­cis­i­on per­for­mance for lea­der­ship AI trai­ning, hel­ping to fuel the con­ver­gence of HPC and AI.

Uti­li­zed in the Fron­tier super­com­pu­ter at Oak Ridge Natio­nal Labo­ra­to­ry, the HPC and AI per­for­mance capa­bi­li­ties in AMD Instinct MI200 series acce­le­ra­tors will be key in enab­ling rese­ar­chers and sci­en­tists to acce­le­ra­te their time to sci­ence and discovery.

“Zen 4” Powered Data Cen­ter, Desi­gned for Lea­der­ship Per­for­mance [31:52 – 36:22]
AMD pro­vi­ded new details on the expan­ded next gene­ra­ti­on AMD EPYC pro­ces­sors code­na­med “Gen­oa” and “Ber­ga­mo.”

“Gen­oa” is expec­ted to be the world’s hig­hest per­for­mance pro­ces­sor for gene­ral pur­po­se com­pu­ting. It will have up to 96 high-per­for­mance “Zen 4” cores pro­du­ced on opti­mi­zed 5nm tech­no­lo­gy, and will sup­port the next gene­ra­ti­on of memo­ry and I/O tech­no­lo­gies with DDR5 and PCIe® 5. “Gen­oa” will also include sup­port for CXL, enab­ling signi­fi­cant memo­ry expan­si­on capa­bi­li­ties for data cen­ter appli­ca­ti­ons. “Gen­oa” is on track for pro­duc­tion and launch in 2022.
“Ber­ga­mo” is a high-core count CPU, tail­or made for cloud nati­ve appli­ca­ti­ons, fea­turing 128 high per­for­mance “Zen 4c” cores. AMD opti­mi­zed the new “Zen 4c” core for cloud-nati­ve com­pu­ting, tuning the core design for den­si­ty and increased power effi­ci­en­cy to enable hig­her core count pro­ces­sors with breakth­rough per­for­mance per-socket. “Ber­ga­mo” comes with all the same soft­ware and secu­ri­ty fea­tures and is socket com­pa­ti­ble with “Gen­oa.” “Ber­ga­mo” is on track to ship in the first half of 2023.

You can watch the full video here and learn more about all of the pro­ducts dis­cus­sed during the event here.

Sup­port­ing Resources

• Learn more about the AMD Instinct™ accelerators
• Learn more about AMD EPYC™ processors 
• Learn more about the Oak Ridge Natio­nal Laboratory’s Fron­tier supercomputer
• Beco­me a fan of AMD on Face­book
• Fol­low AMD on Twit­ter
• Con­nect with AMD On Lin­ke­dIn

About AMD
For more than 50 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: AMD)  web­site, blog, Face­book, Lin­ke­dIn and Twit­ter pages. 

AMD, the AMD Arrow logo, EPYC, AMD Instinct, Infi­ni­ty Fabric and com­bi­na­ti­ons the­reof are trade­marks of Advan­ced Micro Devices, Inc. PCIe is a regis­tered trade­mark of PCI-SIG Cor­po­ra­ti­on. Other names are for infor­ma­tio­nal pur­po­ses only and may be trade­marks of their respec­ti­ve owners.

This press release con­ta­ins for­ward-loo­king state­ments con­cer­ning Advan­ced Micro Devices, Inc. (AMD) such as the fea­tures, func­tion­a­li­ty, per­for­mance, avai­la­bi­li­ty, timing and expec­ted bene­fits of AMD pro­ducts inclu­ding the 3rd Gen AMD EPYC™ pro­ces­sors with AMD 3D V‑Cache and AMD Instinct™ MI200 series acce­le­ra­tors as well as the “Zen 4” data cen­ter road­map, 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,” “may,” “expects,” “belie­ves,” “plans,” “intends,” “pro­jects” and other terms with simi­lar mea­ning. Inves­tors are cau­tio­ned that the for­ward-loo­king state­ments in this press release are based on cur­rent beliefs, assump­ti­ons and expec­ta­ti­ons, speak only as of the date of this press release 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; glo­bal eco­no­mic uncer­tain­ty; loss of a signi­fi­cant cus­to­mer; impact of the COVID-19 pan­de­mic on AMD’s busi­ness, finan­cial con­di­ti­on and results of ope­ra­ti­ons; com­pe­ti­ti­ve mar­kets in which AMD’s pro­ducts are sold; mar­ket con­di­ti­ons of the indus­tries in which AMD pro­ducts are sold; cycli­cal natu­re of the semi­con­duc­tor indus­try; quar­ter­ly and sea­so­nal sales pat­terns; AMD’s abili­ty to ade­qua­te­ly pro­tect its tech­no­lo­gy or other intellec­tu­al pro­per­ty; unfa­vorable cur­ren­cy exch­an­ge rate fluc­tua­tions; abili­ty of third par­ty manu­fac­tu­r­ers to manu­fac­tu­re AMD’s pro­ducts on a time­ly basis in suf­fi­ci­ent quan­ti­ties and using com­pe­ti­ti­ve tech­no­lo­gies; avai­la­bi­li­ty of essen­ti­al equip­ment, mate­ri­als, sub­stra­tes or manu­fac­tu­ring pro­ces­ses; abili­ty to achie­ve expec­ted manu­fac­tu­ring yields for AMD’s pro­ducts; AMD’s abili­ty to intro­du­ce pro­ducts on a time­ly basis with expec­ted fea­tures and per­for­mance levels; AMD’s abili­ty to gene­ra­te reve­nue from its semi-cus­tom SoC pro­ducts; poten­ti­al secu­ri­ty vul­nerabi­li­ties; poten­ti­al secu­ri­ty inci­dents inclu­ding IT outa­ges, data loss, data brea­ches and cyber-attacks; uncer­tain­ties invol­ving the orde­ring and ship­ment of AMD’s pro­ducts; AMD’s reli­ance on third-par­ty intellec­tu­al pro­per­ty to design and intro­du­ce new pro­ducts in a time­ly man­ner; AMD’s reli­ance on third-par­ty com­pa­nies for design, manu­fac­tu­re and sup­p­ly of mother­boards, soft­ware and other com­pu­ter plat­form com­pon­ents; AMD’s reli­ance on Micro­soft and other soft­ware ven­dors’ sup­port to design and deve­lop soft­ware to run on AMD’s pro­ducts; AMD’s reli­ance on third-par­ty dis­tri­bu­tors and add-in-board part­ners; impact of modi­fi­ca­ti­on or inter­rup­ti­on of AMD’s inter­nal busi­ness pro­ces­ses and infor­ma­ti­on sys­tems; com­pa­ti­bi­li­ty of AMD’s pro­ducts with some or all indus­try-stan­dard soft­ware and hard­ware; cos­ts rela­ted to defec­ti­ve pro­ducts; effi­ci­en­cy of AMD’s sup­p­ly chain; AMD’s abili­ty to rely on third par­ty sup­p­ly-chain logi­stics func­tions; AMD’s abili­ty to effec­tively con­trol sales of its pro­ducts on the gray mar­ket; impact of govern­ment actions and regu­la­ti­ons such as export admi­nis­tra­ti­on regu­la­ti­ons, tariffs and trade pro­tec­tion mea­su­res; AMD’s abili­ty to rea­li­ze its defer­red tax assets; poten­ti­al tax lia­bi­li­ties; cur­rent and future claims and liti­ga­ti­on; impact of envi­ron­men­tal laws, con­flict mine­rals-rela­ted pro­vi­si­ons and other laws or regu­la­ti­ons; impact of acqui­si­ti­ons, joint ven­tures and/or invest­ments on AMD’s busi­ness, inclu­ding the announ­ced acqui­si­ti­on of Xilinx, and abili­ty to inte­gra­te acqui­red busi­nesses; AMD’s abili­ty to com­ple­te the Xilinx mer­ger; impact of the announce­ment and pen­den­cy of the Xilinx mer­ger on AMD’s busi­ness; impact of any impair­ment of the com­bi­ned company’s assets on the com­bi­ned company’s finan­cial posi­ti­on and results of ope­ra­ti­on; rest­ric­tions impo­sed by agree­ments gover­ning AMD’s notes and the revol­ving cre­dit faci­li­ty; AMD’s indeb­ted­ness; AMD’s abili­ty to gene­ra­te suf­fi­ci­ent cash to meet its working capi­tal requi­re­ments or gene­ra­te suf­fi­ci­ent reve­nue and ope­ra­ting cash flow to make all of its plan­ned R&D or stra­te­gic invest­ments; poli­ti­cal, legal, eco­no­mic risks and natu­ral dis­as­ters; future impairm­ents of good­will and tech­no­lo­gy licen­se purcha­ses; AMD’s abili­ty to attract and retain qua­li­fied per­son­nel; AMD’s stock pri­ce vola­ti­li­ty; and world­wi­de poli­ti­cal con­di­ti­ons. 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 most recent reports on Forms 10‑K and 10‑Q.

i MI200-01: World’s fas­test data cen­ter GPU is the AMD Instinct™ MI250X. Cal­cu­la­ti­ons con­duc­ted by AMD Per­for­mance Labs as of Sep 15, 2021, for the AMD Instinct™ MI250X (128GB HBM2e OAM modu­le) acce­le­ra­tor at 1,700 MHz peak boost engi­ne clock resul­ted in 95.7 TFLOPS peak theo­re­ti­cal dou­ble pre­cis­i­on (FP64 Matrix), 47.9 TFLOPS peak theo­re­ti­cal dou­ble pre­cis­i­on (FP64), 95.7 TFLOPS peak theo­re­ti­cal sin­gle pre­cis­i­on matrix (FP32 Matrix), 47.9 TFLOPS peak theo­re­ti­cal sin­gle pre­cis­i­on (FP32), 383.0 TFLOPS peak theo­re­ti­cal half pre­cis­i­on (FP16), and 383.0 TFLOPS peak theo­re­ti­cal Bfloat16 for­mat pre­cis­i­on (BF16) floa­ting-point per­for­mance. Cal­cu­la­ti­ons con­duc­ted by AMD Per­for­mance Labs as of Sep 18, 2020 for the AMD Instinct™ MI100 (32GB HBM2 PCIe® card) acce­le­ra­tor at 1,502 MHz peak boost engi­ne clock resul­ted in 11.54 TFLOPS peak theo­re­ti­cal dou­ble pre­cis­i­on (FP64), 46.1 TFLOPS peak theo­re­ti­cal sin­gle pre­cis­i­on matrix (FP32), 23.1 TFLOPS peak theo­re­ti­cal sin­gle pre­cis­i­on (FP32), 184.6 TFLOPS peak theo­re­ti­cal half pre­cis­i­on (FP16) floa­ting-point per­for­mance. Published results on the NVi­dia Ampere A100 (80GB) GPU acce­le­ra­tor, boost engi­ne clock of 1410 MHz, resul­ted in 19.5 TFLOPS peak dou­ble pre­cis­i­on ten­sor cores (FP64 Ten­sor Core), 9.7 TFLOPS peak dou­ble pre­cis­i­on (FP64). 19.5 TFLOPS peak sin­gle pre­cis­i­on (FP32), 78 TFLOPS peak half pre­cis­i­on (FP16), 312 TFLOPS peak half pre­cis­i­on (FP16 Ten­sor Flow), 39 TFLOPS peak Bfloat 16 (BF16), 312 TFLOPS peak Bfloat16 for­mat pre­cis­i­on (BF16 Ten­sor Flow), theo­re­ti­cal floa­ting-point per­for­mance. The TF32 data for­mat is not IEEE com­pli­ant and not included in this com­pa­ri­son. https://www.nvidia.com/content/dam/en-zz/Solutions/Data-Center/nvidia-ampere-architecture-whitepaper.pdf, page 15, Table 1.
ii MLNX-021R: AMD inter­nal test­ing as of 09/27/2021 on 2x 64C 3rd Gen EPYC with AMD 3D V‑Cache (Milan‑X) com­pared to 2x 64C AMD 3rd Gen EPYC 7763 CPUs using cumu­la­ti­ve avera­ge of each of the fol­lo­wing benchmark’s maxi­mum test result score: ANSYS® Flu­ent® 2021.1, ANSYS® CFX® 2021.R2, and Alta­ir Radioss 2021. Results may vary.
iii MI200-31: As of Octo­ber 20th, 2021, the AMD Instinct™ MI200 series acce­le­ra­tors are the “Most advan­ced ser­ver acce­le­ra­tors (GPUs) for data cen­ter,” defi­ned as the only ser­ver acce­le­ra­tors to use the advan­ced 6nm manu­fac­tu­ring tech­no­lo­gy on a ser­ver. AMD on 6nm for AMD Instinct MI200 series ser­ver acce­le­ra­tors. Nvi­dia on 7nm for Nvi­dia Ampere A100 GPU. https://developer.nvidia.com/blog/nvidia-ampere-architecture-in-depth/
iv MI200-02: Cal­cu­la­ti­ons con­duc­ted by AMD Per­for­mance Labs as of Sep 15, 2021, for the AMD Instinct™ MI250X acce­le­ra­tor (128GB HBM2e OAM modu­le) at 1,700 MHz peak boost engi­ne clock resul­ted in 95.7 TFLOPS peak dou­ble pre­cis­i­on matrix (FP64 Matrix) theo­re­ti­cal, floa­ting-point per­for­mance. Published results on the NVi­dia Ampere A100 (80GB) GPU acce­le­ra­tor resul­ted in 19.5 TFLOPS peak dou­ble pre­cis­i­on (FP64 Ten­sor Core) theo­re­ti­cal, floa­ting-point per­for­mance. Results found at: https://www.nvidia.com/content/dam/en-zz/Solutions/Data-Center/nvidia-ampere-architecture-whitepaper.pdf, page 15, Table 1.