Xilinx Alveo U50 Data Center Accelerator Card - GPU computing processor - Alveo U50 - 8 GB

SKU 4XC7A76757
Produktbeskrivelse:Xilinx Alveo U50 Data Center Accelerator Card - GPU beregningsprocessor - Alveo U50 - 8 GB
Enhedstype:GPU beregningsprocessor - lavprofil
Bus Type:PCI Express 3.0 x16 / PCI Express 4.0 x8
Grafik engine:Xilinx Alveo U50
Hukommelse:8 GB HBM2
Blæserløs:Ja
Dimensioner (B x D x H):16.8 cm x 6.9 cm
14-05-2024
  • Specifikationer
  • Beskrivelse
Generelt
Enhedstype:GPU beregningsprocessor - lavprofil - blæserløs
Bus Type:PCI Express 3.0 x16 / PCI Express 4.0 x8
Grafik engine:Xilinx Alveo U50
Egenskaber:UltraScale+ architecture
Hukommelse
Størrelse:8 GB
Teknologi:HBM2
Båndbredde:316 GBps
Diverse
Strømforbrug ved drift:75 Watt
Udstyr inkluderet:Simpelt beslag, beslag i fuld højde
Overensstemmelsesstandarder:C-Tick, GS, FCC Part 15, UL 60950-1 Second Edition, CSA C22.2 No. 60950-1-07 Second Edition, VCCI Class A, MSIP, CNS 13438, IEC 60950-1:2005 Second Edition, EMC 2014/30/EU, LVD 2014/35/EU, IEEE 1588, EN 55032:2015, CISPR 32, IEC 62368-1: 2014 Second Edition, RoHS CNS 15663, EN 55035:2017, AS/NZS CISPR 32, CAN ICES-3(A)/NMB3(A), EN 60950-1:2006+A11:2009+A1:2012+A2:2011+A2:2013
Dybde:16.8 cm
Højde:6.9 cm
Vægt:325 g
Producentgaranti
Service & Support:Begrænset garanti - 1 år
Miljømæssige parametre
Min. driftstemperatur:0 °C
Maks. driftstemperatur:50 °C
Fugtighedsgrad ved brug:8 - 80%
Oplysninger om kompatibilitet
Designet for:Lenovo ThinkSystem SR645 7D2X, 7D2Y ¦ Lenovo ThinkSystem SR665 7D2V, 7D2W
vis mere
The Xilinx Alveo U50 Data Center accelerator cards provide optimized acceleration for workloads in financial computing, machine learning, computational storage, and data search and analytics. Built on Xilinx UltraScale+ architecture and packaged up in an efficient 75-watt, small form factor, and armed with 100 Gbps networking I/O, PCIe Gen4, and HBM, Alveo U50 is designed for deployment in any server.
Alveo accelerator cards are adaptable to changing acceleration requirements and algorithm standards, capable of accelerating any workload without changing hardware, and reduce overall cost of ownership.
  • Faster application performance with HBM memory and PCIe Gen4 interconnect
  • Low latency network capability
  • Accelerates compute, network, storage workloads
  • Maximized application performance as workloads and algorithms evolve through reconfigurable fabric - unlike fixed-architecture alternatives
  • Built for scale out architectures for deployment solutions on the cloud or on-premises interchangeably




De tekniske detaljer kommer fra 3. part. edgemo tager forbehold for fejl.