AMD Showcases Leadership Cloud Performance with New Amazon EC2 Instances Powered by 4th Gen AMD EPYC Processors

볾ma짯zon EC2 M7a ins짯tances offer up to 50 per짯cent more com짯pu짯te per짯for짯mance than M6a ins짯tances and bring an even broa짯der ran짯ge of workloads to AWS

볾ma짯zon EC2 Hpc7a ins짯tances, powered by 4th Gen AMD EPYC pro짯ces짯sors, deli짯ver up to 2.5x bet짯ter per짯for짯mance com짯pared to Ama짯zon EC2 Hpc6a instances

SANTA CLARA, Calif., Aug. 18, 2023 (GLOBE NEWSWIRE) Today, AMD (NASDAQ: AMD) announ짯ced Ama짯zon Web Ser짯vices (AWS) has expan짯ded its 4th Gen AMD EPYC pro짯ces짯sor-based offe짯rings with the gene짯ral avai짯la짯bi짯li짯ty of Ama짯zon Ela짯s짯tic Com짯pu짯te Cloud (EC2) M7a and Ama짯zon EC2 Hpc7a ins짯tances, which offer next-gene짯ra짯ti짯on per짯for짯mance and effi짯ci짯en짯cy for appli짯ca짯ti짯ons that bene짯fit from high per짯for짯mance, high through짯put and tight짯ly cou짯pled HPC workloads, respectively.

For cus짯to짯mers with incre짯asing짯ly com짯plex and com짯pu짯te-inten짯si짯ve workloads, 4th Gen EPYC pro짯ces짯sor-powered Ama짯zon EC2 ins짯tances deli짯ver a dif짯fe짯ren짯tia짯ted offe짯ring for cus짯to짯mers, said David Brown, vice pre짯si짯dent of Ama짯zon EC2 at AWS. 쏞om짯bi짯ned with the power of the AWS Nit짯ro Sys짯tem, both M7a and Hpc7a ins짯tances allow for fast and low-laten짯cy inter짯node com짯mu짯ni짯ca짯ti짯ons, advan짯cing what our cus짯to짯mers can achie짯ve across our gro짯wing fami짯ly of Ama짯zon EC2 instances.

4th Gen EPYC pro짯ces짯sors have unmat짯ched per짯for짯mance lea짯der짯ship; and our col짯la짯bo짯ra짯ti짯on with AWS is deli짯ve짯ring the full 4th Gen EPYC per짯for짯mance per core in new M7a and Hpc7a ins짯tances, said Dan McNa짯ma짯ra, seni짯or vice pre짯si짯dent and gene짯ral mana짯ger, Ser짯ver Busi짯ness Unit, AMD. 쏛s a result, our cus짯to짯mers can do more with less across a wide ran짯ge of workloads, while simul짯ta짯neous짯ly opti짯mi짯zing their businesses.

AMD and AWS Expand EC2 for Grea짯ter Per짯for짯mance and New Applications
Ama짯zon EC2 M7a ins짯tances, first pre짯view짯ed at this year셲 AMD Data Cen짯ter and AI Tech짯no짯lo짯gy Pre짯mie짯re, cater to a wide ran짯ge of workloads, pro짯vi짯ding cus짯to짯mers with up to 50 per짯cent more com짯pu짯te per짯for짯mance than Ama짯zon EC2 M6a ins짯tances, while offe짯ring new pro짯ces짯sor capa짯bi짯li짯ties, such as AVX3-512, VNNI, and BFloat16. In addi짯ti짯on to high per짯for짯mance in gene짯ral pur짯po짯se workloads, the Ama짯zon EC2 M7a ins짯tances are ide짯al for appli짯ca짯ti짯ons that bene짯fit from high per짯for짯mance and high through짯put such as finan짯cial appli짯ca짯ti짯ons, simu짯la짯ti짯on mode짯ling, gam짯ing, mid-sized data stores and more.

Ama짯zon EC2 Hpc7a ins짯tances are desi짯gned for tight짯ly cou짯pled high per짯for짯mance com짯pu짯ting workloads and deli짯ver 2.5x bet짯ter per짯for짯mance com짯pared to Ama짯zon EC2 Hpc6a ins짯tances. The짯se ins짯tances meet the demands of incre짯asing workload com짯ple짯xi짯ty by offe짯ring AWS cus짯to짯mers more com짯pu짯te, memo짯ry and net짯work per짯for짯mance. Powered by 300 Gb/s EFA net짯work band짯width, the Hpc7a ins짯tances are ide짯al짯ly sui짯ted for com짯pu짯te-inten짯si짯ve, laten짯cy-sen짯si짯ti짯ve workloads such as com짯pu짯ta짯tio짯nal flu짯id dyna짯mics, wea짯ther fore짯cas짯ting, mole짯cu짯lar dyna짯mics, and com짯pu짯ta짯tio짯nal che짯mis짯try. Ama짯zon EC2 Hpc7a ins짯tances also bring next-gene짯ra짯ti짯on ser짯ver tech짯no짯lo짯gy like DDR5 memo짯ry, which pro짯vi짯des 50 per짯cent hig짯her memo짯ry band짯width com짯pared to DDR4 memo짯ry to enable high-speed access to data in memory.

To deli짯ver high-level ope짯ra짯tio짯nal intel짯li짯gence for wea짯ther-depen짯dent indus짯tries, DTN deploys a suite of wea짯ther data and models that deli짯ver sophisti짯ca짯ted, high-reso짯lu짯ti짯on out짯puts and requi짯re con짯ti짯nu짯al pro짯ces짯sing of vast amounts of data from inputs across the glo짯be, said Lars Ewe, Chief Pro짯duct and Tech짯no짯lo짯gy Offi짯cer, DTN. 쏰ur col짯la짯bo짯ra짯ti짯on with AWS allows us to bet짯ter ser짯ve our cus짯to짯mers with the most up-to-date wea짯ther intel짯li짯gence that feed tho짯se ana짯ly짯tic engi짯nes. We are exci짯ted to see how the next gene짯ra짯ti짯on of Ama짯zon EC2 Hpc7a ins짯tances can poten짯ti짯al짯ly sup짯port our mis짯si짯on to pro짯vi짯de cus짯to짯mers with the insights they need at the moment they are needed.

AMD began working with AWS in 2018 and now pro짯vi짯des more than 100 EPYC pro짯ces짯sor-based ins짯tances for gene짯ral pur짯po짯se, com짯pu짯te opti짯mi짯zed, memo짯ry opti짯mi짯zed and high per짯for짯mance com짯pu짯ting workloads. Cus짯to짯mers such as DNT, Sprinklr, and True짯Car have all bene짯fit짯ted from signi짯fi짯cant cost and cloud uti짯liza짯ti짯on opti짯miza짯ti짯on with AMD based Ama짯zon EC2 ins짯tances. With the짯se new ins짯tances, AWS cus짯to짯mers can con짯ti짯nue to take advan짯ta짯ge of the excel짯lent per짯for짯mance, sca짯la짯bi짯li짯ty, and effi짯ci짯en짯cy offe짯red by AMD EPYC processors.

Sup짯port짯ing Resources

About AMD
For more than 50 years AMD has dri짯ven inno짯va짯ti짯on in high-per짯for짯mance com짯pu짯ting, gra짯phics and visua짯liza짯ti짯on tech짯no짯lo짯gies. Bil짯li짯ons of peo짯p짯le, lea짯ding For짯tu짯ne 500 busi짯nesses and cut짯ting-edge sci짯en짯ti짯fic rese짯arch insti짯tu짯ti짯ons around the world rely on AMD tech짯no짯lo짯gy dai짯ly to impro짯ve how they live, work and play. AMD employees are focu짯sed on buil짯ding lea짯der짯ship high-per짯for짯mance and adap짯ti짯ve 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: AMDweb짯siteblogLin짯ke짯dIn and Twit짯ter pages.

AMD, the AMD Arrow logo, EPYC and com짯bi짯na짯ti짯ons the짯reof are trade짯marks of Advan짯ced Micro Devices, Inc. Other names are for infor짯ma짯tio짯nal pur짯po짯ses only and may be trade짯marks of their respec짯ti짯ve owners.