AMD Unveils Workload-Tailored Innovations and Products at The Accelerated Data Center Premiere

AMD laun짯ches AMD Instinct꽓 MI200 series acce짯le짯ra짯tors, pre짯views 3rd Gen AMD EPYC꽓 pro짯ces짯sors with AMD 3D V멌ache, and pro짯vi짯des new details on expan짯ded set of next-gene짯ra짯ti짯on EPYC꽓 pro짯ces짯sors powered by 쏾en 4 and 쏾en 4c CPU cores

Meta choo짯ses EPYC꽓 CPUs for its data center 

SANTA CLARA, Calif., Nov. 08, 2021 (GLOBE NEWSWIRE) AMD (NASDAQ: AMD) held the vir짯tu짯al Acce짯le짯ra짯ted Data Cen짯ter Pre짯mie짯re, laun짯ching the new AMD Instinct꽓 MI200 series acce짯le짯ra짯tors, the world셲 fas짯test acce짯le짯ra짯tor for high per짯for짯mance com짯pu짯ting (HPC) and arti짯fi짯ci짯al intel짯li짯gence (AI) workload짯si, and pro짯vi짯ded a pre짯view of the inno짯va짯ti짯ve 3rd Gen AMD EPYC꽓 pro짯ces짯sors with AMD 3D V멌ache. AMD also reve짯a짯led new infor짯ma짯ti짯on about its next gene짯ra짯ti짯on 쏾en 4 pro짯ces짯sor core and announ짯ced the new 쏾en 4c pro짯ces짯sor core, both of which will power future AMD ser짯ver pro짯ces짯sors and are desi짯gned to extend the company셲 lea짯der짯ship pro짯ducts for the data center.

We are in a high-per짯for짯mance com짯pu짯ting mega짯cy짯cle that is dri짯ving demand for more com짯pu짯te to power the ser짯vices and devices that impact every aspect of our dai짯ly lives, said Dr. Lisa Su, pre짯si짯dent and CEO, AMD. 쏻e are buil짯ding signi짯fi짯cant momen짯tum in the data cen짯ter with our lea짯der짯ship pro짯duct port짯fo짯lio, inclu짯ding Meta셲 adop짯ti짯on of AMD EPYC to power their infra짯struc짯tu짯re and the buil짯dout of Fron짯tier, the first U.S. exas짯ca짯le super짯com짯pu짯ter which will be powered by EPYC and AMD Instinct pro짯ces짯sors. In addi짯ti짯on, today we announ짯ced a breadth of new pro짯ducts that build on that momen짯tum in next-gene짯ra짯ti짯on EPYC pro짯ces짯sors with new inno짯va짯tions in design, lea짯der짯ship, 3D pack짯a짯ging tech짯no짯lo짯gy, and 5 nm high-per짯for짯mance manu짯fac짯tu짯ring to fur짯ther extend our lea짯der짯ship for cloud, enter짯pri짯se and HPC customers.

Meta Adopts EPYC CPUs [03:09 05:29]
AMD announ짯ced Meta is the latest major hypers짯ca짯le cloud com짯pa짯ny that has adopted AMD EPYC CPUs to power its data cen짯ters. AMD and Meta work짯ed tog짯e짯ther to defi짯ne an open, cloud-sca짯le, sin짯gle-socket ser짯ver desi짯gned for per짯for짯mance and power effi짯ci짯en짯cy, based on the 3rd Gen EPYC pro짯ces짯sor. Fur짯ther details will be dis짯cus짯sed at the Open Com짯pu짯te Glo짯bal Sum짯mit later this week. 

Advan짯ced Pack짯a짯ging Dri짯ving Data Cen짯ter Per짯for짯mance [05:35 18:00]
AMD pre짯view짯ed the use of inno짯va짯ti짯ve 3D chip짯let pack짯a짯ging tech짯no짯lo짯gy in the data cen짯ter with the first ser짯ver CPU using high-per짯for짯mance 3D die stack짯ing. The 3rd Gen AMD EPYC pro짯ces짯sors with AMD 3D V멌ache, code짯na짯med 쏮ilan멭, repres짯ents an inno짯va짯ti짯ve step for짯ward in CPU design and pack짯a짯ging, and will offer a 50% avera짯ge per짯for짯mance uplift across tar짯ge짯ted tech짯ni짯cal com짯pu짯ting workloadsii.

3rd Gen EPYC with AMD 3D V멌ache will offer the same capa짯bi짯li짯ties and fea짯tures as the 3rd Gen EPYC pro짯ces짯sors and they will be drop-in com짯pa짯ti짯ble with a BIOS upgrade, deli짯ve짯ring easy adop짯ti짯on and per짯for짯mance enhancements.
Micro짯soft Azu짯re HPC vir짯tu짯al machi짯nes fea짯turing 3rd Gen EPYC with AMD 3D V멌ache are available today in Pri짯va짯te Pre짯view, with broad roll짯out in the coming weeks. More infor짯ma짯ti짯on on per짯for짯mance and avai짯la짯bi짯li짯ty is available here.
3rd Gen EPYC CPUs with AMD 3D V멌ache will launch in Q1 2022. Part짯ners inclu짯ding Cis짯co, Dell Tech짯no짯lo짯gies, Leno짯vo, HPE and Super짯mi짯cro are plan짯ning to offer ser짯ver solu짯ti짯ons with the짯se processors.

Deli짯ve짯ring Exas짯ca짯le Class Per짯for짯mance for Acce짯le짯ra짯ted Com짯pu짯ting [18:02 31:50]
AMD laun짯ched the AMD Instinct MI200 series acce짯le짯ra짯tors. Based on the AMD CDNA꽓 2 archi짯tec짯tu짯re, the MI200 series acce짯le짯ra짯tors are the most advan짯ced acce짯le짯ra짯tors in the worl짯diii and pro짯vi짯de up to 4.9x hig짯her peak per짯for짯mance for HPC workload짯siv and 1.2X hig짯her peak flops of mixed pre짯cis짯i짯on per짯for짯mance for lea짯der짯ship AI trai짯ning, hel짯ping to fuel the con짯ver짯gence of HPC and AI.

Uti짯li짯zed in the Fron짯tier super짯com짯pu짯ter at Oak Ridge Natio짯nal Labo짯ra짯to짯ry, the HPC and AI per짯for짯mance capa짯bi짯li짯ties in AMD Instinct MI200 series acce짯le짯ra짯tors will be key in enab짯ling rese짯ar짯chers and sci짯en짯tists to acce짯le짯ra짯te their time to sci짯ence and discovery.

Zen 4 Powered Data Cen짯ter, Desi짯gned for Lea짯der짯ship Per짯for짯mance [31:52 36:22]
AMD pro짯vi짯ded new details on the expan짯ded next gene짯ra짯ti짯on AMD EPYC pro짯ces짯sors code짯na짯med 쏥en짯oa and 쏝er짯ga짯mo.

Gen짯oa is expec짯ted to be the world셲 hig짯hest per짯for짯mance pro짯ces짯sor for gene짯ral pur짯po짯se com짯pu짯ting. It will have up to 96 high-per짯for짯mance 쏾en 4 cores pro짯du짯ced on opti짯mi짯zed 5nm tech짯no짯lo짯gy, and will sup짯port the next gene짯ra짯ti짯on of memo짯ry and I/O tech짯no짯lo짯gies with DDR5 and PCIe짰 5. 쏥en짯oa will also include sup짯port for CXL, enab짯ling signi짯fi짯cant memo짯ry expan짯si짯on capa짯bi짯li짯ties for data cen짯ter appli짯ca짯ti짯ons. 쏥en짯oa is on track for pro짯duc짯tion and launch in 2022.
쏝er짯ga짯mo is a high-core count CPU, tail짯or made for cloud nati짯ve appli짯ca짯ti짯ons, fea짯turing 128 high per짯for짯mance 쏾en 4c cores. AMD opti짯mi짯zed the new 쏾en 4c core for cloud-nati짯ve com짯pu짯ting, tuning the core design for den짯si짯ty and increased power effi짯ci짯en짯cy to enable hig짯her core count pro짯ces짯sors with breakth짯rough per짯for짯mance per-socket. 쏝er짯ga짯mo comes with all the same soft짯ware and secu짯ri짯ty fea짯tures and is socket com짯pa짯ti짯ble with 쏥en짯oa. 쏝er짯ga짯mo 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 pro짯ducts dis짯cus짯sed during the event here.

Sup짯port짯ing Resources

About AMD
For more than 50 years AMD has dri짯ven inno짯va짯ti짯on in high-per짯for짯mance com짯pu짯ting, gra짯phics and visua짯liza짯ti짯on tech짯no짯lo짯gies the buil짯ding blocks for gam짯ing, immersi짯ve plat짯forms and the dat짯a짯cen짯ter. Hundreds of mil짯li짯ons of con짯su짯mers, lea짯ding For짯tu짯ne 500 busi짯nesses and cut짯ting-edge sci짯en짯ti짯fic rese짯arch faci짯li짯ties around the world rely on AMD tech짯no짯lo짯gy dai짯ly to impro짯ve how they live, work and play. AMD employees around the world are focu짯sed on buil짯ding gre짯at pro짯ducts that push the boun짯da짯ries of what is pos짯si짯ble. For more infor짯ma짯ti짯on about how AMD is enab짯ling today and inspi짯ring tomor짯row, visit the AMD (NASDAQ: AMD)  web짯site, blog, Face짯book, Lin짯ke짯dIn and Twit짯ter pages. 

AMD, the AMD Arrow logo, EPYC, AMD Instinct, Infi짯ni짯ty Fabric and com짯bi짯na짯ti짯ons the짯reof are trade짯marks of Advan짯ced Micro Devices, Inc. PCIe is a regis짯tered trade짯mark of PCI-SIG Cor짯po짯ra짯ti짯on. Other names are for infor짯ma짯tio짯nal pur짯po짯ses only and may be trade짯marks of their respec짯ti짯ve owners.

This press release con짯ta짯ins for짯ward-loo짯king state짯ments con짯cer짯ning Advan짯ced Micro Devices, Inc. (AMD) such as the fea짯tures, func짯tion짯a짯li짯ty, per짯for짯mance, avai짯la짯bi짯li짯ty, timing and expec짯ted bene짯fits of AMD pro짯ducts inclu짯ding the 3rd Gen AMD EPYC꽓 pro짯ces짯sors with AMD 3D V멌ache and AMD Instinct꽓 MI200 series acce짯le짯ra짯tors as well as the 쏾en 4 data cen짯ter road짯map, which are made pur짯su짯ant to the Safe Har짯bor pro짯vi짯si짯ons of the Pri짯va짯te Secu짯ri짯ties Liti짯ga짯ti짯on Reform Act of 1995. For짯ward-loo짯king state짯ments are com짯mon짯ly iden짯ti짯fied by words such as 쐗ould, 쐌ay, 쐃xpects, 쐀elie짯ves, 쐏lans, 쐇ntends, 쐏ro짯jects and other terms with simi짯lar mea짯ning. Inves짯tors are cau짯tio짯ned that the for짯ward-loo짯king state짯ments in this press release are based on cur짯rent beliefs, assump짯ti짯ons and expec짯ta짯ti짯ons, speak only as of the date of this press release and invol짯ve risks and uncer짯tain짯ties that could cau짯se actu짯al results to dif짯fer mate짯ri짯al짯ly from cur짯rent expec짯ta짯ti짯ons. Such state짯ments are sub짯ject to cer짯tain known and unknown risks and uncer짯tain짯ties, many of which are dif짯fi짯cult to pre짯dict and gene짯ral짯ly bey짯ond AMD셲 con짯trol, that could cau짯se actu짯al results and other future events to dif짯fer mate짯ri짯al짯ly from tho짯se expres짯sed in, or impli짯ed or pro짯jec짯ted by, the for짯ward-loo짯king infor짯ma짯ti짯on and state짯ments. Mate짯ri짯al fac짯tors that could cau짯se actu짯al results to dif짯fer mate짯ri짯al짯ly from cur짯rent expec짯ta짯ti짯ons include, wit짯hout limi짯ta짯ti짯on, the fol짯lo짯wing: Intel Corporation셲 domi짯nan짯ce of the micro짯pro짯ces짯sor mar짯ket and its aggres짯si짯ve busi짯ness prac짯ti짯ces; glo짯bal eco짯no짯mic uncer짯tain짯ty; loss of a signi짯fi짯cant cus짯to짯mer; impact of the COVID-19 pan짯de짯mic on AMD셲 busi짯ness, finan짯cial con짯di짯ti짯on and results of ope짯ra짯ti짯ons; com짯pe짯ti짯ti짯ve mar짯kets in which AMD셲 pro짯ducts are sold; mar짯ket con짯di짯ti짯ons of the indus짯tries in which AMD pro짯ducts are sold; cycli짯cal natu짯re of the semi짯con짯duc짯tor indus짯try; quar짯ter짯ly and sea짯so짯nal sales pat짯terns; AMD셲 abili짯ty to ade짯qua짯te짯ly pro짯tect its tech짯no짯lo짯gy or other intellec짯tu짯al pro짯per짯ty; unfa짯vorable cur짯ren짯cy exch짯an짯ge rate fluc짯tua짯tions; abili짯ty of third par짯ty manu짯fac짯tu짯r짯ers to manu짯fac짯tu짯re AMD셲 pro짯ducts on a time짯ly basis in suf짯fi짯ci짯ent quan짯ti짯ties and using com짯pe짯ti짯ti짯ve tech짯no짯lo짯gies; avai짯la짯bi짯li짯ty of essen짯ti짯al equip짯ment, mate짯ri짯als, sub짯stra짯tes or manu짯fac짯tu짯ring pro짯ces짯ses; abili짯ty to achie짯ve expec짯ted manu짯fac짯tu짯ring yields for AMD셲 pro짯ducts; AMD셲 abili짯ty to intro짯du짯ce pro짯ducts on a time짯ly basis with expec짯ted fea짯tures and per짯for짯mance levels; AMD셲 abili짯ty to gene짯ra짯te reve짯nue from its semi-cus짯tom SoC pro짯ducts; poten짯ti짯al secu짯ri짯ty vul짯nerabi짯li짯ties; poten짯ti짯al secu짯ri짯ty inci짯dents inclu짯ding IT outa짯ges, data loss, data brea짯ches and cyber-attacks; uncer짯tain짯ties invol짯ving the orde짯ring and ship짯ment of AMD셲 pro짯ducts; AMD셲 reli짯ance on third-par짯ty intellec짯tu짯al pro짯per짯ty to design and intro짯du짯ce new pro짯ducts in a time짯ly man짯ner; AMD셲 reli짯ance on third-par짯ty com짯pa짯nies for design, manu짯fac짯tu짯re and sup짯p짯ly of mother짯boards, soft짯ware and other com짯pu짯ter plat짯form com짯pon짯ents; AMD셲 reli짯ance on Micro짯soft and other soft짯ware ven짯dors sup짯port to design and deve짯lop soft짯ware to run on AMD셲 pro짯ducts; AMD셲 reli짯ance on third-par짯ty dis짯tri짯bu짯tors and add-in-board part짯ners; impact of modi짯fi짯ca짯ti짯on or inter짯rup짯ti짯on of AMD셲 inter짯nal busi짯ness pro짯ces짯ses and infor짯ma짯ti짯on sys짯tems; com짯pa짯ti짯bi짯li짯ty of AMD셲 pro짯ducts with some or all indus짯try-stan짯dard soft짯ware and hard짯ware; cos짯ts rela짯ted to defec짯ti짯ve pro짯ducts; effi짯ci짯en짯cy of AMD셲 sup짯p짯ly chain; AMD셲 abili짯ty to rely on third par짯ty sup짯p짯ly-chain logi짯stics func짯tions; AMD셲 abili짯ty to effec짯tively con짯trol sales of its pro짯ducts on the gray mar짯ket; impact of govern짯ment actions and regu짯la짯ti짯ons such as export admi짯nis짯tra짯ti짯on regu짯la짯ti짯ons, tariffs and trade pro짯tec짯tion mea짯su짯res; AMD셲 abili짯ty to rea짯li짯ze its defer짯red tax assets; poten짯ti짯al tax lia짯bi짯li짯ties; cur짯rent and future claims and liti짯ga짯ti짯on; impact of envi짯ron짯men짯tal laws, con짯flict mine짯rals-rela짯ted pro짯vi짯si짯ons and other laws or regu짯la짯ti짯ons; impact of acqui짯si짯ti짯ons, joint ven짯tures and/or invest짯ments on AMD셲 busi짯ness, inclu짯ding the announ짯ced acqui짯si짯ti짯on of Xilinx, and abili짯ty to inte짯gra짯te acqui짯red busi짯nesses; AMD셲 abili짯ty to com짯ple짯te the Xilinx mer짯ger; impact of the announce짯ment and pen짯den짯cy of the Xilinx mer짯ger on AMD셲 busi짯ness; impact of any impair짯ment of the com짯bi짯ned company셲 assets on the com짯bi짯ned company셲 finan짯cial posi짯ti짯on and results of ope짯ra짯ti짯on; rest짯ric짯tions impo짯sed by agree짯ments gover짯ning AMD셲 notes and the revol짯ving cre짯dit faci짯li짯ty; AMD셲 indeb짯ted짯ness; AMD셲 abili짯ty to gene짯ra짯te suf짯fi짯ci짯ent cash to meet its working capi짯tal requi짯re짯ments or gene짯ra짯te suf짯fi짯ci짯ent reve짯nue and ope짯ra짯ting cash flow to make all of its plan짯ned R&D or stra짯te짯gic invest짯ments; poli짯ti짯cal, legal, eco짯no짯mic risks and natu짯ral dis짯as짯ters; future impairm짯ents of good짯will and tech짯no짯lo짯gy licen짯se purcha짯ses; AMD셲 abili짯ty to attract and retain qua짯li짯fied per짯son짯nel; AMD셲 stock pri짯ce vola짯ti짯li짯ty; and world짯wi짯de poli짯ti짯cal con짯di짯ti짯ons. Inves짯tors are urged to review in detail the risks and uncer짯tain짯ties in AMD셲 Secu짯ri짯ties and Exch짯an짯ge Com짯mis짯si짯on filings, inclu짯ding but not limi짯t짯ed to AMD셲 most recent reports on Forms 10멚 and 10멡.

MI200-01: World셲 fas짯test data cen짯ter GPU is the AMD Instinct꽓 MI250X. Cal짯cu짯la짯ti짯ons con짯duc짯ted by AMD Per짯for짯mance Labs as of Sep 15, 2021, for the AMD Instinct꽓 MI250X (128GB HBM2e OAM modu짯le) acce짯le짯ra짯tor at 1,700 MHz peak boost engi짯ne clock resul짯ted in 95.7 TFLOPS peak theo짯re짯ti짯cal dou짯ble pre짯cis짯i짯on (FP64 Matrix), 47.9 TFLOPS peak theo짯re짯ti짯cal dou짯ble pre짯cis짯i짯on (FP64), 95.7 TFLOPS peak theo짯re짯ti짯cal sin짯gle pre짯cis짯i짯on matrix (FP32 Matrix), 47.9 TFLOPS peak theo짯re짯ti짯cal sin짯gle pre짯cis짯i짯on (FP32), 383.0 TFLOPS peak theo짯re짯ti짯cal half pre짯cis짯i짯on (FP16), and 383.0 TFLOPS peak theo짯re짯ti짯cal Bfloat16 for짯mat pre짯cis짯i짯on (BF16) floa짯ting-point per짯for짯mance. Cal짯cu짯la짯ti짯ons con짯duc짯ted by AMD Per짯for짯mance Labs as of Sep 18, 2020 for the AMD Instinct꽓 MI100 (32GB HBM2 PCIe짰 card) acce짯le짯ra짯tor at 1,502 MHz peak boost engi짯ne clock resul짯ted in 11.54 TFLOPS peak theo짯re짯ti짯cal dou짯ble pre짯cis짯i짯on (FP64), 46.1 TFLOPS peak theo짯re짯ti짯cal sin짯gle pre짯cis짯i짯on matrix (FP32), 23.1 TFLOPS peak theo짯re짯ti짯cal sin짯gle pre짯cis짯i짯on (FP32), 184.6 TFLOPS peak theo짯re짯ti짯cal half pre짯cis짯i짯on (FP16) floa짯ting-point per짯for짯mance. Published results on the NVi짯dia Ampere A100 (80GB) GPU acce짯le짯ra짯tor, boost engi짯ne clock of 1410 MHz, resul짯ted in 19.5 TFLOPS peak dou짯ble pre짯cis짯i짯on ten짯sor cores (FP64 Ten짯sor Core), 9.7 TFLOPS peak dou짯ble pre짯cis짯i짯on (FP64). 19.5 TFLOPS peak sin짯gle pre짯cis짯i짯on (FP32), 78 TFLOPS peak half pre짯cis짯i짯on (FP16), 312 TFLOPS peak half pre짯cis짯i짯on (FP16 Ten짯sor Flow), 39 TFLOPS peak Bfloat 16 (BF16), 312 TFLOPS peak Bfloat16 for짯mat pre짯cis짯i짯on (BF16 Ten짯sor Flow), theo짯re짯ti짯cal floa짯ting-point per짯for짯mance. The TF32 data for짯mat is not IEEE com짯pli짯ant and not included in this com짯pa짯ri짯son. https://www.nvidia.com/content/dam/en-zz/Solutions/Data-Center/nvidia-ampere-architecture-whitepaper.pdf, page 15, Table 1.
ii MLNX-021R: AMD inter짯nal test짯ing as of 09/27/2021 on 2x 64C 3rd Gen EPYC with AMD 3D V멌ache (Milan멭) com짯pared to 2x 64C AMD 3rd Gen EPYC 7763 CPUs using cumu짯la짯ti짯ve avera짯ge of each of the fol짯lo짯wing benchmark셲 maxi짯mum test result score: ANSYS짰 Flu짯ent짰 2021.1, ANSYSCFX짰 2021.R2, and Alta짯ir Radioss 2021. Results may vary.
iii MI200-31: As of Octo짯ber 20th, 2021, the AMD Instinct꽓 MI200 series acce짯le짯ra짯tors are the 쏮ost advan짯ced ser짯ver acce짯le짯ra짯tors (GPUs) for data cen짯ter, defi짯ned as the only ser짯ver acce짯le짯ra짯tors to use the advan짯ced 6nm manu짯fac짯tu짯ring tech짯no짯lo짯gy on a ser짯ver. AMD on 6nm for AMD Instinct MI200 series ser짯ver acce짯le짯ra짯tors. Nvi짯dia on 7nm for Nvi짯dia Ampere A100 GPU. https://developer.nvidia.com/blog/nvidia-ampere-architecture-in-depth/
iv MI200-02: Cal짯cu짯la짯ti짯ons con짯duc짯ted by AMD Per짯for짯mance Labs as of Sep 15, 2021, for the AMD Instinct꽓 MI250X acce짯le짯ra짯tor (128GB HBM2e OAM modu짯le) at 1,700 MHz peak boost engi짯ne clock resul짯ted in 95.7 TFLOPS peak dou짯ble pre짯cis짯i짯on matrix (FP64 Matrix) theo짯re짯ti짯cal, floa짯ting-point per짯for짯mance. Published results on the NVi짯dia Ampere A100 (80GB) GPU acce짯le짯ra짯tor resul짯ted in 19.5 TFLOPS peak dou짯ble pre짯cis짯i짯on (FP64 Ten짯sor Core) theo짯re짯ti짯cal, floa짯ting-point per짯for짯mance. Results found at: https://www.nvidia.com/content/dam/en-zz/Solutions/Data-Center/nvidia-ampere-architecture-whitepaper.pdf, page 15, Table 1.