GeekRadar / Il banco di confronto

CONFRONTA.

GPU

NVIDIA

Tesla V100 FHHL

Processore grafico

Memoria video: 16 GB^*
Bus di memoria: 4.096 bit^*
Banda di memoria nominale: 827,40 GB/s^*

Non sono ancora disponibili benchmark associati a questa scheda.

Elaborazione e piattaforma

Architettura: Volta^*
Nodo produttivo: 12 nm^*
Core GPU: 5.120^*
Frequenza base GPU: 937 MHz^*
Frequenza GPU dichiarata: 937 MHz^*
Interfaccia grafica: PCIe 3.0 x16^*
Area del chip dichiarata: 815 mm²^*
TDP chip/package: 250 W^*
Transistor: 21.100 million^*
Frequenza boost GPU: 1.290 MHz^*
Cache L2 GPU: 6 MB^*
Processore grafico: GV100^*
Unità di texture mapping: 320^*
Unità di rasterizzazione: 128^*
Multiprocessori streaming: 80^*
Tensor core: 640^*
Fonderia del chip: TSMC^*
Formato scheda o modulo: Single-slot^*
TDP GPU dichiarato: 250 W^*
Alimentatore di sistema suggerito: 600 W^*
Connettori di alimentazione esterni: 1x 8-pin^*
DirectX / feature level: 12 (12_1)^*
OpenGL: 4.6^*
OpenCL: 3.0^*
Vulkan: 1.4^*
Compute capability CUDA: 7.0^*
Shader model: 6.9^*

Memoria e archiviazione

Opzioni memoria: 16^*
Memoria video: 16 GB^*
Tipo memoria video: HBM2^*
Bus di memoria: 4.096 bit^*
Banda di memoria nominale: 827,40 GB/s^*

Specifiche di potenza

Potenza termica di progetto: 250 W^*

Configurazione di sistema

Specifiche dettagliate

Specifiche 1 · 11 campi

Manufacturer: Nvidia^*
GFLOPS: 13.210^* GFLOPS
TDP: 250^*
Class: Datacenter^*
Name: NVIDIA Tesla V100 FHHL^*
Year: 2018^*
Fab: 12^*
Transistors (mln): 21.100^*
Die size: 815^*
Memory size: 16.384^*
Memory speed: 829,4^*

Specifiche 2 · 10 campi

Architettura GPU — Tesla V100 FHHL: Volta^*
Supporto hardware nativo: BF16 tensoriale — Tesla V100 FHHL: No^*
Supporto hardware nativo: FP4 tensoriale — Tesla V100 FHHL: No^*
Supporto hardware nativo: FP64 tensoriale — Tesla V100 FHHL: No^*
Supporto hardware nativo: FP6 tensoriale — Tesla V100 FHHL: No^*
Supporto hardware nativo: FP8 tensoriale — Tesla V100 FHHL: No^*
Supporto hardware nativo: INT8 tensoriale — Tesla V100 FHHL: No^*
Supporto hardware nativo: TF32 tensoriale — Tesla V100 FHHL: No^*
Configurazione fisica — Tesla V100 FHHL: FHHL (full height, half length) PCIe^*
Modello GPU — Tesla V100 FHHL: Tesla V100 FHHL^*

Specifiche 3 · 8 campi

Product Name: Tesla V100 FHHL^*
GPU Chip: GV100^*
Released: Mar 27th, 2018^*
Bus: PCIe 3.0 x16^*
Memory: 16 GB, HBM2, 4096 bit^*
GPU clock: 937^* MHz
Memory clock: 808^* MHz
Shaders TMUs ROPs: 5120 / 320 / 128^*

Specifiche 4 · 13 campi

Foundry: TSMC^*
Vendor: NVIDIA^*
FP16 GFLOPS: 26.420^* GFLOPS
FP32 GFLOPS: 13.210^* GFLOPS
FP64 GFLOPS: 6.605^* GFLOPS
Product: NVIDIA Tesla V100 FHHL^*
Type: GPU^*
Release Date: 2018-03-27^*
Process Size (nm): 12^* nm
TDP (W): 250^* W
Die Size (mm^2): 815^* mm²
Transistors (million): 21.100^*
Freq (MHz): 937^* MHz

Specifiche 5 · 30 campi

proc transistors: 21,100 million^*
proc die size: 815^* mm²
cores: 5.120^*
tensor cores: 640^*
sm count: 80^*
base clock: 937^*
boost clock: 1.290^*
memory clock: 808^*
ram: 16^*
name: NVIDIA Tesla V100 FHHL^*
ram type: HBM2^*
bus width: 4.096^*
bus interface: PCIe 3.0 x16^*
ram bandwidth: 827,4^* GB/s
fp16 performance: 26,42^*
fp32 performance: 13,21^*
fp64 performance: 6,605^*
tdp: 250^* W
suggested psu: 600^* W
slot width: Single-slot^*
power connectors: 1x 8-pin^*
api cuda: 7^*
is retail board: No^*
gpi value: 13,41^*
vendor: nvd^*
manufacturer: NVIDIA^*
arch name: Volta^*
card release date: 2018-03-27^*
proc foundry: TSMC^*
proc process size: 12^* nm

Specifiche 6 · 39 campi

Compute capability CUDA

7.0^*

DirectX / feature level

12 (12_1)^*

OpenCL

3.0^*

OpenGL

4.6^*

Shader model

6.9^*

Vulkan

1.4^*

Architettura

Volta^*

Interfaccia grafica

PCIe 3.0 x16^*

Bus di memoria

4.096^* bit

Shader processor dichiarati

5.120^*

Picco teorico FP16 dichiarato

26,42^* TFLOPS

Ambito e condizioni

Densità del calcolo: Non specificata^*

Picco teorico FP32 dichiarato

13,21^* TFLOPS

Ambito e condizioni

Densità del calcolo: Non specificata^*

Picco teorico FP64 dichiarato

6,605^* TFLOPS

Ambito e condizioni

Densità del calcolo: Non specificata^*

Configurazione cache L1 dichiarata

128 KB (per SM)^*

Cache L2 GPU dichiarata

6 MB^*

Uscite video dichiarate

No outputs^*

Fill rate teorico dei pixel

165.1 GPixel/s^*

Connettori di alimentazione esterni

1x 8-pin^*

Area del chip dichiarata

815 mm²^*

Fonderia del chip

TSMC^*

Nodo produttivo dichiarato

12 nm^*

Transistor

21,100 million^*

Processore grafico

GV100^*

Memoria video dichiarata

16^* GB

Banda di memoria nominale

827.4 GB/s^*

Tipo memoria video

HBM2^*

Unità di rasterizzazione

128^*

Unità shader dichiarate

5.120^*

Formato scheda o modulo

Single-slot^*

Multiprocessori streaming

80^*

Alimentatore di sistema suggerito

600 W^*

Sparsità tensoriale 2:4

No^*

TDP GPU dichiarato

250 W^*

Tensor core

640^*

Fill rate teorico delle texture

412.8 GTexel/s^*

Unità di texture mapping

320^*

Core GPU

5.120^*

Memoria video

16^* GB

Tipo memoria video

HBM2^*

Specifiche 7 · 8 campi

Product Name: Tesla V100 FHHL^*
GPU Chip: GV100^*
Released: Mar 27th, 2018^*
Bus: PCIe 3.0 x16^*
Memory: 16 GB, HBM2, 4096 bit^*
GPU clock: 937^* MHz
Memory clock: 808^* MHz
Shaders / TMUs / ROPs: 5120 / 320 / 128^*

Specifiche 8 · 11 campi

Manufacturer: Nvidia^*
GFLOPS: 13.210^* GFLOPS
TDP: 250^*
Class: Datacenter^*
Name: NVIDIA Tesla V100 FHHL^*
Year: 2018^*
Fab: 12^*
Transistors (mln): 21.100^*
Die size: 815^*
Memory size: 16.384^*
Memory speed: 829,4^*

Specifiche 9 · 54 campi

Brand: NVIDIA^*
Top BUS WIDTH: 4.096^* bit
Graphics Processor GPU Name: GV100^*
Graphics Processor Architecture: Volta^*
Render Config SM Count: 80^*
Graphics Features CUDA: 7^*
Render Config Tensor Cores: 640^*
Graphics Processor Process Size: 12^* nm
Graphics Processor Transistors: 21,100 million^*
Graphics Processor Die Size: 815^* mm²
Graphics Card Release Date: Mar 27th, 2018^*
Graphics Card Generation: Tesla Volta (Vxx)^*
Graphics Card Successor: Tesla Turing^*
Graphics Card Production: End-of-life^*
Name: Tesla V100 FHHL^*
Graphics Card Bus Interface: PCIe 3.0 x16^*
Clock Speeds Memory Clock: 808 MHz 1616 Mbps effective^*
Memory Memory Size: 16^* GB
Memory Memory Type: HBM2^*
Memory Memory Bus: 4.096^* bit
Memory Bandwidth: 827,4^* GB/s
Render Config TMUs: 320^*
Top GRAPHICS PROCESSOR: GV100^*
Render Config ROPs: 128^*
Theoretical Performance Pixel Rate: 165.1 GPixel/s
Theoretical Performance Texture Rate: 412.8 GTexel/s
Board Design Slot Width: Single-slot^*
Board Design TDP: 250^* W
Board Design Suggested PSU: 600^* W
Board Design Outputs: No outputs^*
Board Design Board Number: PG502 SKU 200^*
Graphics Features DirectX: 12 (12_1)^*
Graphics Features OpenGL: 4,6^*
Graphics Features OpenCL: 3^*
Graphics Features Vulkan: 1,3^*
source file: nvidia2018.json^*
Graphics Processor Foundry: TSMC^*
Graphics Card Predecessor: Tesla Pascal^*
Board Design Power Connectors: 1x 8-pin^*
Graphics Processor Density: 25.9M / mm²^*
Top TMUS: 320^*
Top ROPS: 128^*
Clock Speeds Base Clock: 937^* MHz
Clock Speeds Boost Clock: 1.290^* MHz
Graphics Features Shader Model: 6.8^*
Top CORES: 5.120^*
Render Config Shading Units: 5.120^*
Render Config L2 Cache: 6^* MB
Top MEMORY SIZE: 16^* GB
Theoretical Performance FP32 (float): 13,21^* TFLOPS
Theoretical Performance FP64 (double): 6.605 TFLOPS (1:2)
Render Config L1 Cache: 128 KB (per SM)^*
Theoretical Performance FP16 (half): 26.42 TFLOPS (2:1)
Top MEMORY TYPE: HBM2^*

Specifiche 10 · 27 campi

manufacturer: NVIDIA^*
manufacturer: NVIDIA^*
rop: 64^*
rop: 128^*
igp: No^*
igp: No^*
bus: PCIe 3.0 x16^*
bus: PCIe 3.0 x16^*
memType: HBM2^*
memType: HBM2^*
gpuChip: Vega 10^*
gpuChip: GV100^*
productName: Tesla V100 FHHL^*
productName: Tesla V100 FHHL^*
releaseYear: 2.018^*
releaseYear: 2.018^*
memSize: 16^*
memSize: 16^*
memBusWidth: 4.096^*
gpuClock: 937^*
gpuClock: 1.138^*
memClock: 810^*
memClock: 800^*
unifiedShader: 5.120^*
unifiedShader: 3.584^*
tmu: 224^*
tmu: 320^*

* Dati provvisori