Description
The PNY NVIDIA A40 48GB GDDR6 (TCSA40M-PB) is a high-performance professional graphics card designed for data centers, enterprise workstations, and AI workloads, leveraging the NVIDIA Ampere architecture (GA102). It features 10,752 CUDA cores, 336 Tensor cores, and 84 RT cores, delivering massive speedups for rendering, simulation, and AI applications.
NVIDIA +4
NVIDIA +4Key Features and Specifications:
- Architecture & Core Count: Based on the 8nm GA102 processor with 10,752 Ampere architecture CUDA cores, 84 second-generation RT Cores, and 336 third-generation Tensor Cores.
- Memory: 48GB GDDR6 ECC (Error Correcting Code) memory, supporting 696 GB/s memory bandwidth.
- Performance: Offers approximately 37.4 TFLOPS of FP32 performance, with up to 5X faster AI training using Tensor Float 32 (TF32) compared to previous generations.
- NVLink Support: Supports 2-Way NVLink with 112.5 GB/s bidirectional bandwidth, allowing two A40 cards to be paired to expand memory up to 96GB.
- Virtualization: Configured for virtualization by default (vGPU), making it ideal for high-end virtual workstations (vWS) accessible remotely.
- Cooling & Power: Features a passive thermal solution suitable for server environments, with a maximum power consumption of 300W (single 8-pin CPU power connector).
- Form Factor & Interface: Dual-slot, full-height card with PCIe 4.0 x16 interface.
NVIDIA +7
Target Workloads:
- AI Development/Data Science: Tensor cores and large memory enable fast AI training, inference, and data science workloads.
- Professional Visualization: Supports NVIDIA RTX Virtual Workstation software for CAD, CAE, and immersive rendering.
- Broadcast/AR/VR: Provides high-resolution display capabilities, live broadcast support, and high-performance virtual reality at the edge.
NVIDIA +1
The A40 is designed for 24/7 data center operations and includes secure and measured boot with hardware root of trust.
PNY.com +1
PNY.com +1
