Nvidia TCSV100MPCIE-PB Product Data Sheet
NVIDIA TESLA V100
1.5X HPC
Perf
orman
c
e in One
Y
ear
CPU System
:
2X Xeon E5-2690v4 @ 2.6GHz
on pre-production hardware
cuFFT
Seismic(RTM)Physics(QUDA)STREAM
0
1.0
X
2.0
X
r
f
rce
omz
t
o P0
0
| GPU System: NVIDIA®Tesla®P100 or V10
0 | V100 measured
8X V10
0
8X P10
0
8X K80
44 Hours
18 Hours
7.4 Hour
s
Deep Learning
Training
in One Workday
Time to Soluti
on in Hour
s
- Lower i
s Bette
r
40502030100
Server Config: Dual Xeon E5-2699 v4, 2.6GHz |
8x Tesla K80, Tesla P100 or Tesla V100 | V100
performance measured on pre-production
hardware. | ResNet-50 Training on Microsoft
Cognitive Toolkit for 90 Epoc
hs with 1.28M
ImageNet dataset
Workload: ResNet-5
0 | CPU: 2X Xeon E5-2690v
4 @ 2.6GHz | GPU: add 1
X
NVIDI
A®Tesla
®
P100 or V100 at 150
W | V100 measured on pre-production hardware
.
Tesla V100
Tesla P1002X CP
U
0
20X40X10X30
X
Performanc
e Normalized to CPU
30x Higher Th
r
oughput than CPU
Server on Deep Learning In
ferenc
e
GPU ACCELERATOR
The Most Advanced Data Center GPU Ever Built.
NVIDIA® Tesla® V100 is the world’s most advanced data center
GPU ever built to accelerate AI, HPC, and graphics. Powered by
NVIDIA Volta™, the latest GPU architecture, Tesla V100 offers the
performance of up to 100 CPUs in a single GPU—enabling data
scientists, researchers, and engineers to tackle challenges that
were once thought impossible.
SPECIFICATIONS
Part Number TCSV100MPCIE-PB
EAN 3536403359041
GPU Architecture NVIDIA Volta
NVIDIA Tensor Cores 640
NVIDIA CUDA® Cores 5,120
Double-Precision
Performance
Single-Precision
Performance
Tensor Performance 112 TFLOPS
GPU Memory 16 GB HBM2
Memory Bandwidth 900 GB/sec
ECC Yes
System Interface PCIe Gen3
Form Factor PCIe Full Height
Max Power
Comsumption
Thermal Solution Passive
Compute APIs CUDA, DirectCompute,
7 TFLOPS
14 TFLOPS
250 W
OpenCL™, OpenACC
GROUNDBREAKING INNOVATIONS
VOLTA ARCHITECTURE
By pairing CUDA Cores and
Tensor Cores within a unified
architecture, a single server
with Tesla V100 GPUs can
replace hundreds of commodity
CPU servers for traditional HPC
and Deep Learning.
MAXIMUM
EFFICIENCY MODE
The new maximum efficiency
mode allows data centers
to achieve up to 40% higher
compute capacity per rack
within the existing power
budget. In this mode, Tesla
V100 runs at peak processing
efficiency, providing up to 80%
of the performance at half the
power consumption.
TENSOR CORE
Equipped with 640 Tensor
Cores, Tesla V100 delivers 120
TeraFLOPS of deep learning
performance. That’s 12X Tensor
FLOPS for DL Training, and 6X
Tensor FLOPS for DL Inference
when compared to NVIDIA
Pascal™ GPUs.
HBM2
With a combination of improved
raw bandwidth of 900 GB/s
and higher DRAM utilization
efficiency at 95%, Tesla V100
delivers 1.5X higher memory
bandwidth over Pascal GPUs
as measured on STREAM.
NEXT GENERATION NVLINK
NVIDIA NVLink in Tesla V100
delivers 2X higher throughput
compared to the previous
generation. Up to eight Tesla
V100 accelerators can be
interconnected at up to 300
GB/s to unleash the highest
application performance
possible on a single server.
PROGRAMMABILITY
Tesla V100 is architected
from the ground up to
C
simplify programmability.
Its new independent thread
scheduling enables finer-grain
synchronization and improves
GPU utilization by sharing
resources among small jobs.
Tesla V100 is the flagship product of Tesla data center computing platform for deep learning, HPC, and graphics.
The Tesla platform accelerates over 450 HPC applications and every major deep learning framework. It is available
everywhere from desktops to servers to cloud services, delivering both dramatic performance gains and cost
savings opportunities.
© 2017 NVIDIA Corporation and PNY. All rights reserved. NVIDIA, the NVIDIA logo,
Quadro, nView, CUDA, NVIDIA Pascal, and 3D Vision are trademarks and/or registered
trademarks of NVIDIA Corporation in the U.S. and other countries. e PNY logotype is
a registered trademark of PNY Technologies. OpenCL is a trademark of Apple Inc. used
under license to the Khronos Group Inc. All other trademarks and copyrights are the
property of their respective owners.
PNY Technologies Europe
Rue Joseph Cugnot BP40181 - 33708 Mérignac Cedex|France
T +33 (0)5 56 13 75 75 | F +33 (0)5 56 13 75 77
For more information visit: www.pny.eu
Loading...