PNY NVIDIA A40 48GB GDDR6 Passive Heatsink Deep Learning Graphics Card Accelerator 699-2G133-0200-C00 / NVA40TCGPU-KIT

ENGINE ARCHITECTURE: AMPERE
CUDA CORES: 10, 752
TENSOR CORES: 336 ( 3RD GEN )
RT CORES: 84 ( 2ND GEN )
GPU CLOCK (
BASE ): 1305 MHZ
BOOST CLOCK: 1740 MHZ
PEAK FP32 TFLOPS PERFORMANCE (
NON TENSOR ): 37.4 ( TERAFLOPS )
PEAK FP16 TENSOR TFOPS WITH FP16 ACCUMULATE PERFORMANCE: 149.7 TFLOPS / 299.4 TFLOPS ( TERAFLOPS )
PEAK TF32 TENSOR PERFORMANCE: 74.8 TFLOPS / 169.6 TFLOPS ( TERAFLOPS )
RT CORE PERFORMANCE: 73.1 TFLOPS ( TERAFLOPS )
PEAK BF16 TENSOR TFLOPS WITH FP32 ACCUMULATE: 149.7 TFLOPS ( TERAFLOPS ) / 299.4 TFLOPS ( TERAFLOPS )
INTEGER PERFORMANCE: INT8 299.3 TOPS / 598.6 TOPS; INT4 598.7 TOPS / 1197.4 TOPS
NVLINK TRANSFER SPEEDS: 112.5GB /S ( BI-DIRECTIONAL )
MEMORY:

MEMORY CONFIGURATION: 48GB
MEMORY INTERFACE: GDDR6
MEMORY INTERFACE WIDTH: 384 BIT
MEMORY BANDWIDTH: 696GB/S
MEMORY CLOCK: 7251 MHZ
ERROR CORRECTION CODE (
ECC ): YES ( ENABLED BY DEFAULT / CAN BE DISABLED VIA SOFTWARE )
SUPPORT:

BUS SUPPORT: PCI-E 4.0
PHYSICAL BUS INTERFACE: PCI-E 4.0 X16
SUPPORTED TECHNOLOGIES: NVIDIA CUDA CORE TECHNOLOGY, DIRECTCOMPUTE, OPENCL, OPENACC, DIRECTX 12.07, SHADER MODEL 5.17, OPENGL 4.68, VULKAN 1.18, LEVEL 3 NEBS READY, NVIDIA VP / VAPPS SUPPORTED, NVIDIA RTX VIRTUAL WORKSTATION SUPPORTED, NVIDIA VIRTUAL COMPUTE SERVER SUPPORTED
OPERATING SYSTEMS: MICROSOFT WINDOWS 2016, MICROSOFT WINDOWS SERVER 2019, MICROSOFT WINDOWS SERVER 2008 R2, MICROSOFT WINDOWS SERVER 2012 R2, LINUX ( ENGLISH US / UK )
CONNECTIONS:

DISPLAY CONNECTORS: ( 3 ) THREE DISPLAYPORTS DP
POWER CONNECTORS: ( 1 ) ONE 8 PIN AUXILIARY POWER CONNECTOR
ADDITIONAL: ( 1 ) ONE NVLINK INTERFACE ( 3RD GEN )
THERMAL AND POWER SPECS:

MAX POWER CONSUMPTION (
W ): 300W
COOLING SYSTEM:

PASSIVE HEATSINK ( BIDIRECTIONAL AIRFLOW )
FORM FACTOR:

DUAL SLOT, FULL HEIGHT / HIGH PROFILE
DIMENSIONS:

4.4 INCH ( H ) ( MEASUREMENT INCLUDES PCIE INTERFACE ) X 10.5 INCH ( L )