New AMD Radeon™ Pro V340 Graphics Card Delivers Accelerated Performance and High User Density to Power Datacenter Visualization Workloads

 
— Harnes­ses the industry’s only hard­ware-based GPU vir­tua­liza­ti­on solu­ti­on enab­led by SR-IOV and the powerful “Vega” archi­tec­tu­re to acce­le­ra­te and secu­re modern visua­liza­ti­on workloads —
 

VMworld, LAS VEGAS, Aug. 26, 2018  — AMD today announ­ced the Rade­on™ Pro V340 gra­phics card, a high-per­for­mance dual-GPU Vir­tu­al Desk­top Infra­struc­tu­re (VDI) solu­ti­on pur­po­se-built to power and acce­le­ra­te the most deman­ding dat­a­cen­ter visua­liza­ti­on workloads, inclu­ding CAD, design, Desk­top as a Ser­vice (DaaS) and rendering.

The AMD Rade­on™ Pro V340 gra­phics card is a dual-GPU solu­ti­on based on the advan­ced AMD “Vega” archi­tec­tu­re1, opti­mi­zed to deli­ver extre­me per­for­mance and high user den­si­ty for vir­tua­li­zed envi­ron­ments. It is the first VDI hard­ware solu­ti­on equip­ped with 32GB of ultra-fast, second-gene­ra­ti­on high-band­width memo­ry, pro­vi­ding mas­si­ve amounts of memo­ry and band­width for today’s most com­plex design and media workloads.

The AMD Rade­on™ Pro V340 gra­phics card is enab­led by AMD MxG­PU Tech­no­lo­gy, the industry’s only hard­ware-based GPU vir­tua­liza­ti­on solu­ti­on, which is based on the indus­try-stan­dard SR-IOV (Sin­gle Root I/O Vir­tua­liza­ti­on) tech­no­lo­gy. Com­bi­ning soft­ware and hard­ware tech­no­lo­gies that deli­ver vir­tua­li­zed gra­phics for the modern cloud, AMD MxG­PU deli­vers fast, sta­ble and pre­dic­ta­ble per­for­mance with the industry’s hig­hest user den­si­ty2, wit­hout requi­ring recur­ring end-user licen­se fees. 

As the flag­ship of our new Rade­on™ Pro V‑series pro­duct line, the Rade­on™ Pro V340 gra­phics card employs advan­ced secu­ri­ty fea­tures and helps to cost effec­tively deli­ver and acce­le­ra­te modern visua­liza­ti­on workloads from the dat­a­cen­ter,” said Ogi Brkic, gene­ral mana­ger of Rade­on Pro at AMD.

The AMD Rade­on™ Pro V340 gra­phics card will enable our cus­to­mers to secu­re­ly levera­ge desk­top and appli­ca­ti­on vir­tua­liza­ti­on for the most gra­phi­cal­ly deman­ding appli­ca­ti­ons,” said Shel­don D’Paiva, direc­tor of Pro­duct Mar­ke­ting at VMware. “With Rade­on™ Pro for VMware, admins can easi­ly set up a VDI envi­ron­ment, rapidly deploy vir­tu­al GPUs to exis­ting vir­tu­al machi­nes and enable hundreds of pro­fes­sio­nals with just a few mou­se clicks.”

With increased den­si­ty, fas­ter frame buf­fer and enhan­ced secu­ri­ty, the AMD Rade­on™ Pro V340 gra­phics card deli­vers a powerful new choice for our cus­to­mers to power their Citrix Workspace, even for the most deman­ding appli­ca­ti­ons,” said Cal­vin Hsu, VP of Pro­duct Mar­ke­ting at Citrix.

Pur­po­se-built Technology

The AMD Rade­on™ Pro V340 gra­phics card deli­vers advan­ced fea­tures and tech­no­lo­gies geared towards enter­pri­se, DaaS and cloud gam­ing solu­ti­ons to acce­le­ra­te visua­liza­ti­on workloads, including:

  • Supe­ri­or User Den­si­ty: Sup­ports up to 32 1GB vir­tu­al machi­nes, up to 33 per­cent more than the com­pe­ti­ti­ve solu­ti­on.2
  • Inte­gra­ted Encode Engi­ne: The abili­ty to com­press inde­pen­dent video streams in both H.264 and H.265 for­mats. Pro­vi­des design and manu­fac­tu­ring users with the video qua­li­ty they expect, while empowe­ring IT mana­gers to eli­mi­na­te CPU bottlenecks.
  • Ultra-Fast Frame Buf­fer: HBM2 memo­ry with Error Cor­rec­ting Code (ECC)3and bet­ter power effi­ci­en­cy than com­pe­ting solu­ti­ons allow fast paging apps to run incre­di­bly fast.
  • Built-in Secu­ri­ty Pro­ces­sor: Pro­vi­des secu­re boot and encrypt­ed sto­rage capabilities.

Avai­la­bi­li­ty

The AMD Rade­on™ Pro V340 gra­phics card is expec­ted to be available in Q4 2018 from lea­ding sys­tem pro­vi­ders. Visit the AMD booth #2501 at VMworld in Las Vegas to see the new solu­ti­on in action.

Sup­port­ing Resources

About AMD
For more than 45 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:AMDweb­siteblogFace­book and Twit­ter pages.

1 The infor­ma­ti­on con­tai­ned her­ein is for infor­ma­tio­nal pur­po­ses only, and is sub­ject to chan­ge wit­hout noti­ce. Time­lines, road­maps, and/or pro­duct release dates shown her­ein are plans only and sub­ject to chan­ge. “Vega” and “Pola­ris” are code­na­mes for AMD archi­tec­tures, and are not pro­duct names. GD-122

2 The AMD Rade­on™ Pro V340 gra­phics card sup­ports up to 32 vir­tu­al users with 1GB memo­ry pro­files per card.  The NVIDIA Tes­la P40 sup­ports up to 24 vir­tu­al users with 1GB memo­ry pro­files per card. AMD user den­si­ty advan­ta­ge 32/24 = up to 33% grea­ter. NVIDIA Source:
https://images.nvidia.com/content/pdf/grid/data-sheet/nvidia-p40-datasheet.pdf   

3 ECC sup­port is limi­t­ed to the HBM2 memo­ry and ECC pro­tec­tion is not pro­vi­ded for inter­nal GPU structures.

©2018 Advan­ced Micro Devices, Inc.  All rights reser­ved. AMD, the AMD Arrow logo, Rade­on, and com­bi­na­ti­ons the­reof are trade­marks of Advan­ced Micro Devices, Inc. (AMD) OpenGL® and the oval logo are trade­marks or regis­tered trade­marks of Hew­lett Packard Enter­pri­se in the United Sta­tes and/or other count­ries world­wi­de. DirectX® is a regis­tered trade­mark of Micro­soft Cor­po­ra­ti­on in the US and other juris­dic­tions. Open­CL™ is a trade­mark of Apple Inc. used by per­mis­si­on by Khro­nos Group, Inc. Other pro­duct names used in this publi­ca­ti­on are for iden­ti­fi­ca­ti­on pur­po­ses only and may be trade­marks of their respec­ti­ve companies.