|
gptkbp:instanceOf
|
gptkb:graphics_card
|
|
gptkbp:architecture
|
parallel computing
|
|
gptkbp:codename
|
gptkb:Volta
|
|
gptkbp:firstReleased
|
2017
|
|
gptkbp:hasConcept
|
gptkb:GV100
|
|
https://www.w3.org/2000/01/rdf-schema#label
|
Nvidia Volta GPU
|
|
gptkbp:manufacturer
|
gptkb:Nvidia
|
|
gptkbp:market
|
gptkb:artificial_intelligence
high-performance computing
datacenter
|
|
gptkbp:memoryType
|
gptkb:HBM2
|
|
gptkbp:notableFeature
|
gptkb:NVLink_interconnect
high memory bandwidth
dynamic parallelism
improved energy efficiency
dedicated Tensor Cores for AI
hardware acceleration for deep learning
large L2 cache
|
|
gptkbp:notableModel
|
gptkb:Titan_V
gptkb:Quadro_GV100
gptkb:Tesla_V100
|
|
gptkbp:notebookSupport
|
no
|
|
gptkbp:predecessor
|
Nvidia Pascal GPU
|
|
gptkbp:processNode
|
12nm
|
|
gptkbp:size
|
815 mm² (GV100)
|
|
gptkbp:successor
|
Nvidia Turing GPU
|
|
gptkbp:supports
|
gptkb:OpenGL_4.6
gptkb:CUDA
gptkb:NVLink
gptkb:PCIe_3.0
gptkb:DirectX_12
gptkb:Vulkan_1.1
gptkb:OpenCL_1.2
gptkb:Tensor_Cores
FP16
FP32
FP64
Shader Model 6.1
|
|
gptkbp:transistorCount
|
21.1 billion (GV100)
|
|
gptkbp:usedIn
|
gptkb:Nvidia_Tesla_V100
gptkb:Nvidia_Titan_V
|
|
gptkbp:bfsParent
|
gptkb:Nvidia_Drive_AGX
|
|
gptkbp:bfsLayer
|
7
|