Kategorie: News

AMD Processors Accelerating Performance of Top Supercomputers Worldwide

Verfasst vonpipin

Growing preference for EPYC™ processors resulted in the number of AMD-powered supercomputers growing 3.5x year-over-year

SANTA CLARA, Calif., Nov. 16, 2021 (GLOBE NEWSWIRE) — During this year’s Super­com­pu­ting Con­fe­rence 2021 (SC21), AMD (NASDAQ: AMD) is show­ca­sing its expan­ded pre­sence and gro­wing pre­fe­rence in the high per­for­mance com­pu­ting (HPC) indus­try with the excep­tio­nal inno­va­ti­on and adop­ti­on of AMD data cen­ter pro­ces­sors and acce­le­ra­tors. Cus­to­mers across the indus­try con­ti­nue to expand their use of AMD EPYC™ pro­ces­sors and AMD Instinct™ acce­le­ra­tors to power cut­ting-edge rese­arch nee­ded to address some of the world’s big­gest chal­lenges in cli­ma­te, life sci­en­ces, medi­ci­ne, and more.

Gro­wing pre­fe­rence for AMD is show­ca­sed in the latest Top500 list. AMD now powers 73 super­com­pu­ters, com­pared to 21 in the Novem­ber 2020 list, a more than 3x year-over-year increase. Addi­tio­nal­ly, AMD powers four out of the top ten most powerful super­com­pu­ters in the world, as well as the most powerful super­com­pu­ter in EMEA. Final­ly, AMD EPYC 7003 series pro­ces­sors, which laun­ched eight months ago, are uti­li­zed by 17 of the 75 AMD powered super­com­pu­ters in the list, demons­t­ra­ting the rapid adop­ti­on of the latest gene­ra­ti­on of EPYC processors.

“The demands of super­com­pu­ting users have increased expo­nen­ti­al­ly as the world seeks to acce­le­ra­te rese­arch, redu­cing the time to dis­co­very of valuable infor­ma­ti­on,” said For­rest Nor­rod, seni­or vice pre­si­dent and gene­ral mana­ger, Data Cen­ter and Embedded Solu­ti­ons Busi­ness Group, AMD. “With AMD EPYC CPUs and Instinct acce­le­ra­tors, we con­ti­nue to evol­ve our pro­duct offe­ring to push the boun­da­ries of data cen­ter tech­no­lo­gies enab­ling fas­ter rese­arch, bet­ter out­co­mes and more impact on the world.”

AMD has also been reco­gni­zed in the annu­al HPC­wire Rea­ders’ and Edi­tors’ Choice Awards at SC21. The com­pa­ny won ten awards inclu­ding Best Sus­taina­bi­li­ty Inno­va­ti­on in HPC, Best HPC Ser­ver Pro­duct and the Out­stan­ding Lea­der­ship in HPC award pre­sen­ted to Pre­si­dent and CEO Dr. Lisa Su. (…) Wei­ter­le­sen »

Still waiting for Exascale: Japan’s Fugaku outperforms all competition once again

Verfasst vonpipin

FRANKFURT, Ger­ma­ny; BERKELEY, Calif.; and KNOXVILLE, Tenn.— The 58th annu­al edi­ti­on of the TOP500 saw litt­le chan­ge in the Top10. The Micro­soft Azu­re sys­tem cal­led Voy­a­ger-EUS2 was the only machi­ne to shake up the top spots, clai­ming No. 10. Based on an AMD EPYC pro­ces­sor with 48 cores and 2.45GHz working tog­e­ther with an NVIDIA A100 GPU and 80 GB of memo­ry, Voy­a­ger-EUS2 also uti­li­zes a Mel­lan­ox HDR Infi­ni­band for data transfer. 

While the­re were no other chan­ges to the posi­ti­ons of the sys­tems in the Top10, Perl­mut­ter at NERSC impro­ved its per­for­mance to 70.9 Pflop/s. Housed at the Law­rence Ber­ke­ley Natio­nal Labo­ra­to­ry, Perlmutter’s increased per­for­mance couldn’t move it from its pre­vious­ly held No. 5 spot. 

Fug­a­ku con­ti­nues to hold the No. 1 posi­ti­on that it first ear­ned in June 2020. Its HPL bench­mark score is 442 Pflop/s, which excee­ded the per­for­mance of Sum­mit at No. 2 by 3x. Instal­led at the Riken Cen­ter for Com­pu­ta­tio­nal Sci­ence (R‑CCS) in Kobe, Japan, it was co-deve­lo­ped by Riken and Fuji­tsu and is based on Fujitsu’s cus­tom ARM A64FX pro­ces­sor. Fug­a­ku also uses Fujitsu’s Tofu D inter­con­nect to trans­fer data bet­ween nodes. 

In sin­gle or fur­ther-redu­ced pre­cis­i­on, which are often used in machi­ne lear­ning and A.I. appli­ca­ti­on, Fug­a­ku has a peak per­for­mance abo­ve 1,000 PFlop/s (1 Exaflop/s). As a result, Fug­a­ku is often intro­du­ced as the first “Exas­ca­le” super­com­pu­ter. (…) Wei­ter­le­sen »

Xilinx Launches Alveo U55C, Its Most Powerful Accelerator Card Ever, Purpose-Built for HPC and Big Data Workloads

Verfasst vonpipin

Breakthrough HPC clustering solution and simplified programmability enable massive scale-out of cutting-edge compute across existing customer infrastructure and network

ST. LOUIS–(BUSINESS WIRE)–SC21 – Xilinx, Inc. (NASDAQ: XLNX), the lea­der in adap­ti­ve com­pu­ting, today at the SC21 super­com­pu­ting con­fe­rence intro­du­ced the Alveo™ U55C data cen­ter acce­le­ra­tor card and a new stan­dards-based, API-dri­ven clus­te­ring solu­ti­on for deploy­ing FPGAs at mas­si­ve sca­le. The Alveo U55C acce­le­ra­tor brings supe­ri­or per­for­mance-per-watt to high per­for­mance com­pu­ting (HPC) and data­ba­se workloads and easi­ly sca­les through the Xilinx® HPC clus­te­ring solution.

Pur­po­se-built for HPC and big data workloads, the new Alveo U55C card is the company’s most powerful Alveo acce­le­ra­tor card ever, offe­ring the hig­hest com­pu­te den­si­ty and HBM capa­ci­ty in the Alveo acce­le­ra­tor port­fo­lio. Tog­e­ther with the new Xilinx RoCE v2-based clus­te­ring solu­ti­on, a broad spec­trum of cus­to­mers with lar­ge-sca­le com­pu­te workloads can now imple­ment powerful FPGA-based HPC clus­te­ring using their exis­ting data cen­ter infra­struc­tu­re and net­work. (…) Wei­ter­le­sen »

Samsung Develops Industry’s First LPDDR5X DRAM

Verfasst vonpipin

Samsung’s LPDDR5X DRAM will provide over 1.3x faster processing speeds and consumer nearly 20% less power than the previous LPDDR5 solution

The LPDDR5X solution will broaden the use of high-performance, low-power memory beyond smartphones to AI and edge applications

 

Sam­sung Elec­tro­nics Co., Ltd., the world lea­der in advan­ced memo­ry tech­no­lo­gy, today announ­ced that it has deve­lo­ped the industry’s first 14-nano­me­ter (nm) based 16-giga­bit (Gb) Low Power Dou­ble Data Rate 5X (LPDDR5X) DRAM, desi­gned to dri­ve fur­ther growth throug­hout the high-speed data ser­vice appli­ca­ti­ons inclu­ding 5G, arti­fi­ci­al intel­li­gence (AI) and the metaverse.
(…) Wei­ter­le­sen »

Neue Downloads der KW 45

Verfasst vonpipin

In den letz­ten Woche gab es mal wie­der eine Aktua­li­sie­rung des AMD-Gra­fik­trei­bers (21.11.2), der die Per­for­mance von Batt­le­field 2042 für RX 6000er Kar­ten ver­bes­sert hat. Zusätz­lich wur­de der AMDGPU All-Open und AMDG­PU-Pro Trei­ber für Linux in der Ver­si­on 21.40.1 freigegeben.

Neu auf­ge­nom­men haben wir AMD Store­MI das in der Ver­si­on 2.01.205 erschie­nen ist. Mit dem Tool kön­nen die Start- und Lade­zei­ten von einer HDD mit­hil­fe einer SSD beschleu­nigt wer­den. Ein Caching-Sys­tem spie­gelt dabei die Daten auf die SSD. Vor­aus­set­zung hier­für ist ein Main­board mit AMD X570, PRO 565, B550, A520, 400 Series, X399, TRX40 oder WRX80.

Neue Versionen

Neu aufgenommene Programme

(…) Wei­ter­le­sen »

AMD — Präsentation Accelerated Data Center Premiere

Verfasst vonpipin

Für die AMD-Ver­an­stal­tung vom letz­ten Mon­tag, der “Acce­le­ra­ted Data Cen­ter Pre­mie­re”, rei­chen wir hier­mit die Prä­sen­ta­ti­on nach, die vor­ab als Infor­ma­ti­on her­aus­ge­ge­ben wur­de. Dar­in sind aller­dings nicht die Updates der AMD-Road­map bzw. die Daten zu Zen 4 ent­hal­ten. (…) Wei­ter­le­sen »

AMD beseitigte zahlreiche Sicherheitslücken im Grafiktreiber für Windows 10

Verfasst vonpipin

In einem in die­ser Woche erschie­ne­nen Secu­ri­ty Bul­le­tin hat AMD ins­ge­samt 27 Sicher­heits­lü­cken im Gra­fik­trei­ber für Win­dows 10 auf­ge­führt, von denen man die meis­ten bereits im Jahr 2020 besei­tigt hat. Zwei davon wur­den aller­dings erst mit der Ver­si­on 21.4.1 ange­gan­gen und besit­zen jeweils die Ein­stu­fung “Hoch”.

Es emp­fiehlt sich also nicht nur wegen der bei­den letzt­ge­nann­ten Sicher­heits­lü­cken, CVE-2020–12960 und CVE-2020–12981, oder zum Bei­spiel wegen der Per­for­mance­ver­bes­se­run­gen im gestern erschie­ne­nen Rade­on Soft­ware Adre­na­lin 21.11.2 den Gra­fik­trei­ber aktu­ell zu halten.
(…) Wei­ter­le­sen »

Avnet Collaborates with Digital Realty to Develop “Try and Buy” High Performance, Cloud-Based Universal Video Streaming Solutions

Verfasst vonpipin

Solu­ti­ons enable com­pa­nies to be their own video broad­cas­ters and appli­ca­ti­on pro­vi­ders to enhan­ce their pro­duct offering

PHOENIX–Leading glo­bal tech­no­lo­gy solu­ti­ons pro­vi­der Avnet (Nasdaq: AVT) is team­ing up with Digi­tal Real­ty (NYSE: DLR), the lar­gest glo­bal pro­vi­der of cloud- and car­ri­er-neu­tral data cen­ter, colo­ca­ti­on and inter­con­nec­tion solu­ti­ons, to equip com­pa­nies and appli­ca­ti­on pro­vi­ders to deve­lop, launch, sca­le and/or enhan­ce their video broad­cast pro­duct offe­rings. Joint­ly deve­lo­ped spe­ci­fi­cal­ly for high per­for­mance, cloud-based and secu­re uni­ver­sal video strea­ming, the solu­ti­ons from Digi­tal Real­ty and Avnet will be one-touch (tran­sac­tion), cost-effec­ti­ve “try and buy” opti­ons. The team’s first available solu­ti­on uti­li­zes a 3rd Gen AMD EPYC-based ser­ver inte­gra­ted with a Xilinx U30 FPGA* acce­le­ra­tor card deli­ve­ring excep­tio­nal per­for­mance and flexibility.

A recent stu­dy by Deloit­te (2021) notes that the avera­ge Ame­ri­can home has four digi­tal video ser­vice sub­scrip­ti­ons and the same stu­dy found that 39% of Baby Boo­mers and 29% of Gene­ra­ti­on X rank on demand video as their pre­fer­red enter­tain­ment acti­vi­ty. The joint Avnet and Digi­tal Real­ty video strea­ming solu­ti­ons enable sup­port for a wide base of video strea­ming appli­ca­ti­ons such as: video-based eCom­mer­ce, social media-based appli­ca­ti­ons, video col­la­bo­ra­ti­on, tele­me­di­ci­ne, live sports and events, auc­tions, gambling and sports bet­ting. (…) Wei­ter­le­sen »

ASUS rüstet Renoir- und Vermeer-Support bei A320-Mainboard nach

Verfasst vonNero24

Kürz­lich mach­te in der Gerüch­te­kü­che die Run­de, dass eini­ge Main­board-Her­stel­ler geneigt sein könn­ten, aktu­el­le CPUs nach­träg­lich doch noch für ihre alten Main­boards zu unter­stüt­zen, um die Halb­lei­ter­knapp­heit an die­ser Stel­le etwas zu ent­zer­ren. Ob das der Grund für den Umschwenk war, ist nicht bekannt, jedoch ist Leser tom1tom im Forum auf­ge­fal­len, dass ASUS für das Prime A320M‑K ein neu­es BIOS ver­öf­fent­lich hat, das es in sich hat. (…) Wei­ter­le­sen »

Weiterer Windows-11-Patch behebt Ryzen-Leistungsverlust nicht vollständig

Verfasst vonNero24

Ende Okto­ber haben sowohl Micro­soft als auch AMD einen Fix bzw. einen neu­en Chip­satz-Trei­ber ver­öf­fent­licht und nach AMDs aktu­el­ler Sta­tus-Info gilt das Pro­blem damit als aus der Welt geschafft. Mes­sun­gen der L3-Cache-Per­for­mance nach den Fixes bestä­tig­ten zwar, dass die Cache-Laten­zen wie­der auf nor­ma­lem Niveau sind, der Durch­satz jedoch liegt nach wie vor unter den Wer­ten von Win­dows 10. (…) Wei­ter­le­sen »

CXL™ Consortium & Gen‑Z Consortium Sign Letter of Intent to Advance Interconnect Technology

Verfasst vonpipin

Novem­ber 10, 2021 — High per­for­mance com­pu­ting con­ti­nues to evolve—meeting the ever-incre­asing demand for high effi­ci­en­cy, low-laten­cy, rapid and seam­less pro­ces­sing. The Gen‑Z Con­sor­ti­um was foun­ded in 2016 to crea­te a next-gene­ra­ti­on fabric capa­ble of bridging exis­ting solu­ti­ons while enab­ling new, unboun­ded inno­va­ti­on in an open, non-pro­prie­ta­ry stan­dards body.

In 2019, the CXL™ Con­sor­ti­um laun­ched to deli­ver Com­pu­te Express Link™ (CXL™), an indus­try-sup­port­ed cache-coher­ent inter­con­nect desi­gned for pro­ces­sors, memo­ry expan­si­on, and acce­le­ra­tors. The CXL Con­sor­ti­um and the Gen‑Z Con­sor­ti­um estab­lished a joint memo­ran­dum of under­stan­ding (MOU) pro­vi­ding an oppor­tu­ni­ty for col­la­bo­ra­ti­on to defi­ne bridging bet­ween the pro­to­cols. This took the form of a joint working group that encou­ra­ged crea­ti­vi­ty and inno­va­ti­on bet­ween the two orga­niza­ti­ons toward the bet­ter­ment of the indus­try as a whole. 

Loo­king to the future, the CXL Con­sor­ti­um and Gen‑Z Con­sor­ti­um have iden­ti­fied syn­er­gies bet­ween the two con­sor­tia that resul­ted in the sig­ning of a Let­ter of Intent which, if pas­sed and agreed upon by all par­ties, would trans­fer the Gen‑Z Spe­ci­fi­ca­ti­ons and all Gen‑Z assets to the CXL Con­sor­ti­um. (…) Wei­ter­le­sen »

IBM Cloud Selects 3rd Gen AMD EPYC™ Processors for New Bare Metal Offering for Compute-Intensive Workloads

Verfasst vonpipin

SANTA CLARA, Calif. — 11/10/2021- AMD (NASDAQ: AMD) announ­ced today that IBM Cloud has cho­sen 3rd Gen AMD EPYC™ pro­ces­sors to expand its bare metal ser­vice offe­rings desi­gned to power cus­to­mers’ deman­ding workloads and solu­ti­ons. The new ser­vers, fea­turing 128 cores, up to 4TB of memo­ry and 10 NVMe dri­ves per ser­ver, give users full access to high-end, dual-socket per­for­mance with AMD EPYC 7763 pro­ces­sors; a first for IBM Cloud in a dual-socket platform.

“Our cus­to­mers have a high demand for com­pu­ting pro­ces­sing power and the new 3rd Gen AMD EPYC pro­ces­sors pro­vi­de the high levels of per­for­mance and sca­la­bi­li­ty we were loo­king for,” said Suresh Gopa­l­a­krish­n­an, vice pre­si­dent, IBM Cloud. “Our col­la­bo­ra­ti­on with AMD has hel­ped us deli­ver our hig­hest core counts and band­width ever available for IBM Cloud cus­to­mers, to offer top mar­ket per­for­mance for today and tomorrow’s deman­ding workloads.” (…) Wei­ter­le­sen »

AMD EPYC Milan‑X mit 3D-V-Cache zuerst in Azure-Cloud

Verfasst vonNero24

Auf dem ange­kün­dig­ten Dat­a­cen­ter-Event, das inter­es­sier­te Zuse­her gestern Abend auch bei uns im Forum ver­fol­gen konn­ten, hat AMD unter ande­rem die neu­en EPYC Ser­ver-Pro­zes­so­ren mit Code­na­men Milan‑X und 3D-V-Cache vor­ge­stellt. Es han­delt sich dabei um eine Tech­no­lo­gie, die AMD bereits im Som­mer anhand eines spe­zi­el­len Ryzen 5900X demons­triert hat­te. Im Fal­le des EPYC mit sei­nen bis zu acht CCDs ergibt dies eine Gesamt-L3-Cache-Grö­ße von sagen­haf­ten 768 MB. (…) Wei­ter­le­sen »

DATEV-Ausfall stellt Steuerkanzleien vor Probleme

Verfasst vonNero24

Der gest­ri­ge Mon­tag, 8. Novem­ber 2021, wird DATEV-Kun­den wohl noch lan­ge in Erin­ne­rung blei­ben. Die DATEV ist ein Soft­ware­haus bzw. IT-Dienst­leis­ter vor­wie­gend für Steu­er­be­ra­ter, Wirt­schafts­prü­fer und Rechts­an­wäl­te, die Buch­hal­tun­gen und Löh­ne ihrer Man­dan­ten über die Soft­ware respek­ti­ve Platt­form abwi­ckeln. Das war jedoch gestern nicht mehr mög­lich, ab ca. 10 Uhr Vor­mit­tag ging im DATEV-Rechen­zen­trum nichts mehr. Ver­schärft wur­de die Situa­tio­nen dadurch, dass Unter­neh­men bis zum 10. des Monats ihre Umsatz­steu­er­vor­anmel­dun­gen abge­ge­ben haben müs­sen. (…) Wei­ter­le­sen »

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

Verfasst vonpipin

— 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.” (…) Wei­ter­le­sen »