Kategorie: Pressemitteilungen

In die­ser Sek­ti­on wer­den Pres­se­mit­tei­lun­gen der Her­stel­ler 1:1 ver­öf­fent­licht. Pla­net 3DNow! über­nimmt weder für die Rich­tig­keit der Anga­ben noch für die Art der For­mu­lie­rung jed­we­de Ver­ant­wor­tung. Die Anga­ben kön­nen sub­jek­tiv sein und geben nicht not­wen­di­ger­wei­se die Mei­nung der auf Pla­net 3DNow! schrei­ben­den Autoren wieder.

ASUS AIO-Kühler werden kompatibel zu AMD AM5 Mainboards

Verfasst vonpipin

Die All-In-One-Küh­lun­gen ROG Ryu­jin, Ryuo und Strix LC unter­stüt­zen mit dem neu­en AM5-Mon­ta­ge-Kit den kom­men­den AMD-CPU-Sockel, wäh­rend TUF Gam­ing LC-Küh­ler bereits jetzt kom­pa­ti­bel sind

Ratin­gen, Deutsch­land, 20. Juni 2022 — ASUS gibt heu­te bekannt, dass aus­ge­wähl­te All-in-One-Küh­ler (AIO) voll­stän­dig mit dem AM5-Sockel (LGA 1718) kom­pa­ti­bel sind, der für AMD-CPUs der nächs­ten Gene­ra­ti­on ent­wi­ckelt wur­de. Die ROG Ryujin‑, Ryuo- und Strix-LC-Küh­ler wer­den mit­hil­fe eines neu­en AM5-Mon­ta­ge­kits* umge­rüs­tet, wäh­rend die TUF Gam­ing-LC-Küh­ler mit den bestehen­den Kits ver­wen­det wer­den kön­nen. Eine Lis­te der kom­pa­ti­blen ASUS-Küh­ler fin­den Sie in der fol­gen­den Tabel­le. (…) Wei­ter­le­sen »

VIPER Gaming gibt die Verfügbarkeit der Venom 6200 MHz DDR5 Speichermodule bekannt

Verfasst vonpipin

Rot­ter­dam, Nie­der­lan­de, 9. Juni 2022 — Viper™, geschütz­te Mar­ke von Patri­ot™, einem renom­mier­ten Anbie­ter von Spei­cher­mo­du­len, SSD und Flash-Spei­cher­lö­sun­gen sowie Gam­ing Peri­phe­rie, freut sich, die Ver­füg­bar­keit der VIPER VENOM RGB- und Non-RGB-DDR5-Spei­cher­kits mit 6200 MHz bekannt zu geben. Die Venom DDR5 Modu­le kom­men in Dual-Chan­nel-Kit-Kapa­zi­tä­ten von 16 GB und  32 GB in den Han­del, um die hohen Anfor­de­run­gen von Enthu­si­as­ten und Gamern zu erfüllen.


(…) Wei­ter­le­sen »

World’s Top Ten IC Design Company Revenue Reached US$39.43 billion in 1Q22, Marvell Growth Rate Tops List, Says TrendForce

Verfasst vonpipin

Accor­ding to the latest Trend­Force sta­tis­tics, the top ten IC desi­gners world­wi­de pos­ted a com­bi­ned reve­nue of US$39.43 bil­li­on in 1Q22, or 44% growth YoY. Qual­comm, NVIDIA, Broad­com ran­ked in the top three. After the acqui­si­ti­on of Xilinx, AMD sur­pas­sed Media­Tek in the fourth posi­ti­on. In addi­ti­on, accor­ding to Trend­Force track­ing of IC design indus­try trends, reve­nue gene­ra­ted by Will Semi­con­duc­tor and Cir­rus Logic was enough to be included in the top ten for the first time.

Bene­fiting from growth per­for­mance in hand­sets and RF front-end divi­si­ons in addi­ti­on to its IoT and auto­mo­ti­ve divi­si­ons in 1Q22, Qualcomm’s quar­ter­ly reve­nue rea­ched US$9.55 bil­li­on, or 52% growth YoY, ran­king num­ber one in the world. The expan­ded appli­ca­ti­on of GPUs in data cen­ters boos­ted this por­ti­on of NVIDIA’s reve­nue to 45.4%, sur­pas­sing the 45% accoun­ted for by its gam­ing busi­ness, com­bi­ning for a total reve­nue of US$7.9 bil­li­on, or 53% growth YoY. Broadcom’s reve­nue from semi­con­duc­tor solu­ti­ons is sub­stan­ti­al, inclu­ding net­work chips, broad­band com­mu­ni­ca­ti­on chips, and sto­rage and bridging chips. Its busi­ness has main­tai­ned sta­ble sales per­for­mance, with reve­nue rea­ching US$6.11 bil­li­on, or 26% growth YoY. After the addi­ti­on of Xilinx, AMD’s reve­nue rea­ched US$5.89 bil­li­on, or 71% growth YoY. Howe­ver, even exclu­ding Xilinx, due to strong sales in its enter­pri­se, embedded and semi-cus­to­mi­zed divi­si­ons, AMD’s own busi­ness reve­nue still hit an all-time high of US$5.33 billion.
(…) Wei­ter­le­sen »

Erster europäischer Exascale-Superrechner kommt nach Jülich

Verfasst vonpipin

EuroHPC JU, das Euro­pean High Per­for­mance Com­pu­ting Joint Under­ta­king, wählt das For­schungs­zen­trum Jülich als Part­ner im deut­schen Gauss Cent­re for Super­com­pu­ting aus, um den ers­ten euro­päi­schen Super­com­pu­ter der nächs­ten Gene­ra­ti­on zu betreiben.

Jülich, 15. Juni 2022 – Die Ent­schei­dung ist gefal­len. Das For­schungs­zen­trum Jülich wird Stand­ort des ers­ten euro­päi­schen Exas­ca­le-Rech­ners. Der Super­com­pu­ter soll als ers­ter Rech­ner in Euro­pa die Gren­ze von 1 Tril­li­on Rechen­ope­ra­tio­nen pro Sekun­de – einer „1“ mit 18 Nul­len – bre­chen. Beschafft wer­den soll das Sys­tem von der euro­päi­schen Super­com­pu­ting-Initia­ti­ve EuroHPC JU. Der Exas­ca­le-Rech­ner wird dazu bei­tra­gen, bedeu­ten­de und drän­gen­de wis­sen­schaft­li­che Fra­gen zu lösen, etwa zum Kli­ma­wan­del, zur Bewäl­ti­gung von Pan­de­mien und zur nach­hal­ti­gen Ener­gie­er­zeu­gung, und den inten­si­ven Ein­satz von Künst­li­cher Intel­li­genz sowie die Ana­ly­se gro­ßer Daten­men­gen ermög­li­chen. Die Gesamt­kos­ten für das Sys­tem belau­fen sich auf 500 Mio. Euro. 250 Mio. Euro wer­den von der euro­päi­schen Super­com­pu­ting-Initia­ti­ve EuroHPC JU und 250 Mio. Euro zu glei­chen Tei­len vom Bun­des­mi­nis­te­ri­um für Bil­dung und For­schung (BMBF) und dem Minis­te­ri­um für Kul­tur und Wis­sen­schaft des Lan­des Nord­rhein-West­fa­len (MKW NRW) getra­gen. (…) Wei­ter­le­sen »

AMD Details Strategy to Drive Next Phase of Growth Across $300 Billion Market for High-Performance and Adaptive Computing Solutions

Verfasst vonpipin

AMD unveils next-gene­ra­ti­on hard­ware and soft­ware road­maps, expan­ded pro­duct port­fo­lio addres­sing new mar­kets, and stra­te­gies to acce­le­ra­te data cen­ter growth and deli­ver per­va­si­ve AI leadership

SANTA CLARA, Calif., June 09, 2022 (GLOBE NEWSWIRE) — Today at its Finan­cial Ana­lyst Day, AMD (NASDAQ: AMD) out­lined its stra­tegy to deli­ver its next pha­se of growth dri­ven by the company’s expan­ded port­fo­lio of high-per­for­mance and adap­ti­ve com­pu­ting pro­ducts span­ning the data cen­ter, embedded, cli­ent, and gam­ing markets.

“From the cloud and PCs to com­mu­ni­ca­ti­ons and intel­li­gent end­points, AMD’s high-per­for­mance and adap­ti­ve com­pu­ting solu­ti­ons play an incre­asing­ly lar­ger role in sha­ping the capa­bi­li­ties of near­ly every ser­vice and pro­duct defi­ning the future of com­pu­ting today,” said Dr. Lisa Su, AMD chair and CEO. “The clo­se of our trans­for­ma­tio­nal acqui­si­ti­on of Xilinx and our expan­ded port­fo­lio of lea­der­ship com­pu­te engi­nes pro­vi­de AMD with signi­fi­cant oppor­tu­ni­ties to deli­ver con­tin­ued strong reve­nue growth with com­pel­ling share­hol­der returns as we cap­tu­re a lar­ger share of the diver­se $300 bil­li­on mar­ket for our high-per­for­mance and adap­ti­ve pro­ducts.” (…) Wei­ter­le­sen »

AMD Announces Pricing of Senior Notes Offering

Verfasst vonpipin

SANTA CLARA, Calif., June 07, 2022 (GLOBE NEWSWIRE) — Fol­lo­wing the pre­vious announce­ment of the launch of a seni­or notes offe­ring, AMD (NASDAQ: AMD) announ­ced today that it has pri­ced an under­writ­ten public offe­ring (the “Offe­ring”) of $1 bil­li­on aggre­ga­te prin­ci­pal amount of seni­or notes, con­sis­ting of $500 mil­li­on aggre­ga­te prin­ci­pal amount of its 3.924% Seni­or Notes due 2032 (the “2032 Notes”) and $500 mil­li­on aggre­ga­te prin­ci­pal amount of its 4.393% Seni­or Notes due 2052 (the “2052 Notes,” and tog­e­ther with the 2032 Notes, the “Notes”).

The 2032 Notes will matu­re on June 1, 2032 and will bear inte­rest at a rate of 3.924% per annum, and the 2052 Notes will matu­re on June 1, 2052 and will bear inte­rest at a rate of 4.393% per annum. The Notes will be seni­or unse­cu­red obli­ga­ti­ons of the Company.
The Offe­ring is expec­ted to clo­se on June 9, 2022, sub­ject to the satis­fac­tion of cus­to­ma­ry clo­sing con­di­ti­ons. The Com­pa­ny intends to use the net pro­ceeds from this offe­ring for gene­ral cor­po­ra­te purposes.
(…) Wei­ter­le­sen »

Dell — Newest Precision Powerhouse Features AMD Ryzen Threadripper PRO

Verfasst vonpipin

Pre­cis­i­on 7865 Tower unites per­for­mance and com­pu­ting power in AMD-powered Pre­cis­i­on workstation.

It’s no coin­ci­dence that in our 25th anni­ver­sa­ry year of Dell Pre­cis­i­on, we’ve unvei­led some of the most powerful work­sta­tions ever – with no signs of slo­wing. It began ear­lier this year when we intro­du­ced the world’s most powerful 14-inch mobi­le work­sta­tion¹, fol­lo­wed by last month’s unvei­ling of the new Pre­cis­i­on 7000 Series mobi­le work­sta­tions. Not only do the­se offer the latest tech­no­lo­gi­cal advance­ments, the Pre­cis­i­on 7000 Series fea­tures a new form fac­tor for DDR5 memo­ry that we anti­ci­pa­te will beco­me a new indus­try standard.

Which brings us to today, and the intro­duc­tion of the Pre­cis­i­on 7865 Tower, fea­turing the AMD Ryzen™ Thre­ad­rip­per™ PRO 5000 WX-Series pro­ces­sors. Whe­ther you’re crea­ting the next ani­ma­ted block­bus­ter, deve­lo­ping trend-set­ting AR initia­ti­ves or ana­ly­zing mas­si­ve data sets, this pro­fes­sio­nal work­sta­tion brings tog­e­ther immense power, fle­xi­bi­li­ty and relia­bi­li­ty in one system.

(…) Wei­ter­le­sen »

AMD Announces Senior Notes Offering

Verfasst vonpipin

SANTA CLARA, Calif., June 07, 2022 (GLOBE NEWSWIRE) — AMD (NASDAQ: AMD) announ­ced today that it intends to offer, sub­ject to mar­ket and other con­di­ti­ons, seni­or notes (the “Notes”) in an under­writ­ten public offe­ring (the “Offe­ring”). The Notes will be seni­or unse­cu­red obli­ga­ti­ons of the Com­pa­ny. AMD intends to use the net pro­ceeds from this offe­ring for gene­ral cor­po­ra­te purposes.

Bar­clays Capi­tal Inc., BofA Secu­ri­ties, Inc., Cre­dit Suis­se Secu­ri­ties (USA) LLC, J.P. Mor­gan Secu­ri­ties LLC, and Wells Far­go Secu­ri­ties, LLC are acting as joint book-run­ning mana­gers and under­wri­ters for the Offering.

The Offe­ring will be made pur­su­ant to an auto­ma­ti­cal­ly effec­ti­ve shelf regis­tra­ti­on state­ment filed with the U.S. Secu­ri­ties and Exch­an­ge Com­mis­si­on (the “SEC”) on June 6, 2022 (the “Regis­tra­ti­on State­ment”). Copies of the preli­mi­na­ry pro­s­pec­tus sup­ple­ment and the accom­pany­ing base pro­s­pec­tus for the Offe­ring may be obtai­ned by cont­ac­ting Bar­clays Capi­tal Inc. at 1–888-603‑5847 (toll free), BofA Secu­ri­ties, Inc. at 1–800-294‑1322 (toll free), Cre­dit Suis­se Secu­ri­ties (USA) LLC at 1–800-221‑1037 (toll free), J.P. Mor­gan Secu­ri­ties LLC at 1–212-834‑4533 and Wells Far­go Secu­ri­ties, LLC at +1–800-645‑3751 (toll free). Copies of the preli­mi­na­ry pro­s­pec­tus sup­ple­ment and the accom­pany­ing base pro­s­pec­tus will also be available on the SEC’s web­site at http://www.sec.gov. (…) Wei­ter­le­sen »

AMD to Host Financial Analyst Day on June 9, 2022

Verfasst vonpipin

SANTA CLARA, Calif., June 02, 2022 (GLOBE NEWSWIRE) — AMD (NASDAQ: AMD) announ­ced that the com­pa­ny will host its Finan­cial Ana­lyst Day on Thurs­day, June 9, 2022, to share updates to the company’s cor­po­ra­te stra­tegy, tech­no­lo­gy and pro­duct road­maps, long-term finan­cial model tar­gets and growth oppor­tu­ni­ties. A live video web­cast of the pre­sen­ta­ti­on with slides will be available on the Inves­tor Rela­ti­ons por­ti­on of its web­site www.amd.com. A replay of the web­cast can be acces­sed appro­xi­m­ate­ly 4 hours after the con­clu­si­on of the live event, along with copies of exe­cu­ti­ve presentations.

(…) Wei­ter­le­sen »

AMD Expands High Performance Compute Fund to Aid Researchers Solving the World’s Toughest Challenges

Verfasst vonpipin

AMD Combines HPC Fund with Xilinx Heterogeneous Accelerated Compute Cluster (HACC) Program and Adds Seven Petaflops of Supercomputer Capacity to Power Groundbreaking Research

SANTA CLARA, Calif., June 01, 2022 (GLOBE NEWSWIRE) — AMD (NASDAQ: AMD) today announ­ced the expan­si­on of its High Per­for­mance Com­pu­te (HPC) Fund with the addi­ti­on of 7 peta­flops of com­pu­ting power to assist glo­bal rese­ar­chers working to sol­ve the most deman­ding chal­lenges facing socie­ty today. AMD also announ­ced that the AMD HPC Fund will now inte­gra­te Xilinx Hete­ro­ge­neous Acce­le­ra­ted Com­pu­te Clus­ters (HACC) pro­gram, pro­vi­ding rese­ar­chers with access to AMD EPYC™ pro­ces­sors, AMD Instinct™ acce­le­ra­tors, Xilinx Alveo™ acce­le­ra­tors and Xilinx Ver­sal™ ACAPs to advan­ce rese­arch in are­as inclu­ding cli­ma­te chan­ge, health care, trans­por­ta­ti­on, big data and more.

The new con­tri­bu­ti­on brings the total amount of com­pu­ting capa­ci­ty dona­ted by AMD to over 20 peta­flops with a mar­ket value of more than $31 mil­li­on, as of May 2022. The addi­tio­nal com­pu­ting resour­ces will build on the AMD COVID-19 HPC Fund that was estab­lished in 2020 to pro­vi­de rese­arch insti­tu­ti­ons with com­pu­ting resour­ces to acce­le­ra­te medi­cal rese­arch on COVID-19 and other diseases.

(…) Wei­ter­le­sen »

Expandable Embedded System with AMD Ryzen Embedded V2000 Processor

Verfasst vonpipin

Tai­pei, Tai­wan, May 31, 2022 — IBASE Tech­no­lo­gy Inc. (TPEx: 8050), a world lea­der in embedded boards and com­pu­ting solu­ti­ons, unveils its CMI211-989 embedded sys­tem powered by AMD Ryzen Embedded V2000 pro­ces­sors based on the inno­va­ti­ve “Zen 2” x86 core archi­tec­tu­re using advan­ced 7nm manu­fac­tu­ring pro­cess. The sys­tem harnes­ses the power AMD Ryzen™ Embedded V2748/V2718 CPU with Rade­on™ Gra­phics to enable 25% more PCIe lanes and deli­ver 40 per­cent bet­ter gra­phics per­for­mance over the pre­vious gene­ra­ti­on SoC.

CMI211-989 hou­ses IBASE’s MI989 Mini-ITX mother­board sup­port­ing up to four inde­pen­dent 4K dis­plays via DP 1.4 ports to crea­te 3D immersi­ve expe­ri­en­ces in retail, gam­ing, and fac­to­ry auto­ma­ti­on appli­ca­ti­ons. Mea­su­ring 269(w) x 193(d) x 31.5(h) mm, it offers rich I/O con­nec­ti­vi­ty, sto­rage, expan­si­on capa­bi­li­ties and sup­ports 4x COM, 3x USB3.1, 2x USB2.0, a 2.5” dri­ve bay and a PCIe(x16) slot for riser cards expan­si­on. (…) Wei­ter­le­sen »

Ayar Labs to Accelerate Development and Application of Optical Interconnects in Artificial Intelligence/Machine Learning Architectures with NVIDIA

Verfasst vonpipin

SANTA CLARA, Calif. –May 25, 2022 – Ayar Labs, the lea­der in chip-to-chip opti­cal con­nec­ti­vi­ty, is deve­lo­ping with NVIDIA ground­brea­king arti­fi­ci­al intel­li­gence (AI) infra­struc­tu­re based on opti­cal I/O tech­no­lo­gy to meet future demands of AI and high per­for­mance com­pu­ting (HPC) workloads. The col­la­bo­ra­ti­on will focus on inte­gra­ting Ayar Labs’ tech­no­lo­gy to deve­lop sca­le-out archi­tec­tures enab­led by high-band­width, low-laten­cy and ultra-low-power opti­cal-based inter­con­nects for future NVIDIA pro­ducts. Tog­e­ther, the com­pa­nies plan to acce­le­ra­te the deve­lo­p­ment and adop­ti­on of opti­cal I/O tech­no­lo­gy to sup­port the explo­si­ve growth of AI and machi­ne lear­ning (ML) appli­ca­ti­ons and data volu­mes. (…) Wei­ter­le­sen »

World’s First Exascale Supercomputer Powered by AMD EPYC™ Processors and AMD Instinct™ Accelerators

Verfasst vonpipin

— Fron­tier super­com­pu­ter, powered by AMD EPYC CPUs and AMD Instinct Acce­le­ra­tors, achie­ves num­ber one spots on Top500, Green500 and HPL-AI per­for­mance lists, an indus­try first —

— AMD powers five of the top ten most powerful and eight of the top ten most effi­ci­ent super­com­pu­ters in the world —

SANTA CLARA, Calif., May 30, 2022 (GLOBE NEWSWIRE) — Today, AMD (NASDAQ: AMD) show­ca­sed its con­tin­ued lea­der­ship in high per­for­mance com­pu­ting (HPC) by powe­ring the fas­test and the most ener­gy effi­ci­ent super­com­pu­ters in the world, accor­ding to the latest Top500 and Green500 lists.

Oak Ridge Natio­nal Laboratory’s (ORNL) Fron­tier sys­tem sub­mit­ted its very first score to the Top500 list of 1.1 exa­flops, making it the world’s fas­test super­com­pu­ter and the first to break the exas­ca­le bar­ri­er. Frontier’s per­for­mance is more than dou­ble the num­ber two sys­tem and grea­ter than the sum of the next seven sys­tems on the latest Top500 list.

In addi­ti­on, the Fron­tier test and deve­lo­p­ment sys­tem (TDS) secu­red the top spot on the Green500 list, deli­ve­ring 62.68 gigaflops/watt power-effi­ci­en­cy from a sin­gle cabi­net of opti­mi­zed 3rd Gen AMD EPYC™ pro­ces­sors and AMD Instinct™ MI250x acce­le­ra­tors. Final­ly, Frontier’s mixed-pre­cis­i­on com­pu­ting per­for­mance clo­cked in at 6.86 exa­flops, as mea­su­red by the High-Per­for­mance Lin­pack-Acce­le­ra­tor Intro­s­pec­tion, or HPL-AI, test. Next steps for Fron­tier include con­tin­ued test­ing and vali­da­ti­on of the sys­tem, which remains on track for final accep­tance and ear­ly sci­ence access later in 2022, and open for full sci­ence at the begin­ning of 2023. (…) Wei­ter­le­sen »

TYAN Brings Modern HPC Server Platforms for Data Centers at ISC 2022

Verfasst vonpipin

Tai­pei, Tai­wan, May 27, 2022 — TYAN®, an indus­try-lea­ding ser­ver plat­form design manu­fac­tu­rer and a MiT­AC Com­pu­ting Tech­no­lo­gy Cor­po­ra­ti­on sub­si­dia­ry, will exhi­bit its latest HPC plat­forms powered by 3rd Gen AMD EPYC™ Pro­ces­sors with AMD 3D V‑Cache™ tech­no­lo­gy for data cen­ters deploy­ments at ISC 2022 event from May 30th to June 1st at booth#D400 in Ham­burg, Germany.

“The amount of data gene­ra­ted by humans and machi­nes has increased expo­nen­ti­al­ly and this requi­res a ste­ady increase of data cen­ter com­pu­te per­for­mance, the­r­e­fo­re, modern data cen­ters need balan­ced hard­ware to effi­ci­ent­ly mana­ge gro­wing data volu­mes”, said Dan­ny Hsu, Vice Pre­si­dent of MiT­AC Com­pu­ting Tech­no­lo­gy Corporation’s Ser­ver Infra­struc­tu­re Busi­ness Unit. “TYAN’s ser­ver plat­forms built upon AMD EPYC™ 7003 Series pro­ces­sors pro­vi­de the best foun­da­ti­on for dat­a­cen­ters that can be tail­o­red to spe­ci­fic workloads with balan­cing hard­ware design.”

Opti­mi­zing for HPC com­pu­ta­ti­on, data­ba­se per­for­mance, and high data through­put at the edge

Powered by 3rd Gen AMD EPYC pro­ces­sors with AMD 3D V‑Cache tech­no­lo­gy, the Trans­port HX TN83-B8251 is a 2U dual-socket ser­ver with 16 DDR4-3200 DIMM slots and eight 3.5‑inch hot-swap SATA or NVMe U.2 tool-less dri­ve bays. The plat­form sup­ports up to four dou­ble-wide GPU cards and two addi­tio­nal low-pro­fi­le PCIe 4.0 x16 slots that pro­vi­de opti­mi­zed topo­lo­gy to impro­ve HPC and deep lear­ning per­for­mance. (…) Wei­ter­le­sen »

AMD Instinct™ MI200 Adopted for Large-Scale AI Training in Microsoft Azure

Verfasst vonpipin

ꟷ Micro­soft Azu­re is the first public cloud to deploy AMD Instinct MI200 acce­le­ra­tors ꟷ

SANTA CLARA, Calif., May 26, 2022 (GLOBE NEWSWIRE) — AMD (NASDAQ: AMD) and Micro­soft con­tin­ued their col­la­bo­ra­ti­on in the cloud, with Micro­soft announ­cing the use of AMD Instinct™ MI200 acce­le­ra­tors to power lar­ge sca­le AI trai­ning workloads. In addi­ti­on, Micro­soft announ­ced it is working clo­se­ly with the PyTorch Core team and AMD data cen­ter soft­ware team to opti­mi­ze the per­for­mance and deve­lo­per expe­ri­ence for cus­to­mers run­ning PyTorch on Micro­soft Azu­re and ensu­re that deve­lo­pers’ PyTorch pro­jects take advan­ta­ge of the per­for­mance and fea­tures of AMD Instinct accelerators.

“AMD Instinct MI200 acce­le­ra­tors pro­vi­de cus­to­mers with cut­ting-edge AI and HPC per­for­mance and will power many of the world’s fas­test super­com­pu­ter sys­tems that push the boun­da­ries of sci­ence,” said Brad McCre­die, cor­po­ra­te vice pre­si­dent, Data Cen­ter and Acce­le­ra­ted Pro­ces­sing, AMD. “Our work with Micro­soft to enable lar­ge-sca­le AI trai­ning and infe­rence in the cloud not only high­lights the expan­si­ve tech­no­lo­gy col­la­bo­ra­ti­on bet­ween the two com­pa­nies, but the per­for­mance capa­bi­li­ties of AMD Instinct MI200 acce­le­ra­tors and how they will ulti­m­ate­ly help cus­to­mers advan­ce the gro­wing demands for AI workloads.” (…) Wei­ter­le­sen »