|
gptkbp:instance_of
|
gptkb:architecture
|
|
gptkbp:architecture
|
gptkb:NVIDIA
|
|
gptkbp:competes_with
|
gptkb:AMD_Vega_architecture
|
|
gptkbp:designed_by
|
gptkb:NVIDIA
|
|
gptkbp:enhances
|
FP32 performance
FP64 performance
FP16 performance
|
|
gptkbp:features
|
gptkb:Tensor_Cores
|
|
https://www.w3.org/2000/01/rdf-schema#label
|
NVIDIA Volta Architecture
|
|
gptkbp:improves
|
AI performance
Deep Learning performance
High-Performance Computing (HPC) performance
|
|
gptkbp:includes
|
HBM2 memory
|
|
gptkbp:introduced
|
gptkb:NVLink_2.0
|
|
gptkbp:is_based_on
|
NVIDIA Pascal architecture principles
|
|
gptkbp:is_compatible_with
|
gptkb:NVIDIA_RAPIDS_AI
gptkb:NVIDIA_Tensor_RT
gptkb:NVIDIA_cu_DNN
gptkb:NVIDIA_CUDA_Toolkit
gptkb:NVIDIA_Deep_Stream_SDK
|
|
gptkbp:is_known_for
|
Energy efficiency
Flexibility
Scalability
High throughput
|
|
gptkbp:is_optimized_for
|
gptkb:AI_technology
Deep Learning inference
Deep Learning training
|
|
gptkbp:is_part_of
|
gptkb:NVIDIA_GPU_family
|
|
gptkbp:is_supported_by
|
NVIDIA software stack
|
|
gptkbp:is_used_in
|
gptkb:cloud_computing
Natural language processing
Computer vision
Data analytics
Supercomputers
|
|
gptkbp:manufacturer
|
12 nm process
|
|
gptkbp:predecessor
|
gptkb:NVIDIA_Pascal_Architecture
|
|
gptkbp:provides
|
Unified Memory
Multi-instance GPU (MIG) capability
|
|
gptkbp:released_in
|
gptkb:2017
|
|
gptkbp:successor
|
gptkb:NVIDIA_Turing_Architecture
|
|
gptkbp:supports
|
gptkb:virtual_reality
gptkb:CUDA
Ray tracing
HPC workloads
|
|
gptkbp:targets
|
gptkb:Artificial_Intelligence
gptkb:machine_learning
Data centers
Scientific computing
|
|
gptkbp:used_in
|
gptkb:NVIDIA_Tesla_V100
gptkb:NVIDIA_Titan_V
|
|
gptkbp:bfsParent
|
gptkb:NVIDIA_Tesla_V100
|
|
gptkbp:bfsLayer
|
5
|