The NVIDIA® A40 GPU is an evolutionary leap in performance and multi-workload capabilities from the data center, combining best-in-class professional graphics with powerful compute and AI acceleration to meet today’s design, creative, and scientific challenges. Driving the next generation of virtual workstations and server-based workloads, NVIDIA A40 brings state-of-the-art features for ray-traced rendering, simulation, virtual production, and more to professionals anytime, anywhere.
Powered by the NVIDIA Ampere Architecture
NVIDIA Ampere architecture-based products offer significant performance, features, and benefits advantages over prior generation offerings including:
NVIDIA Ampere Architecture CUDA Cores
Double-speed processing for single-precision floating point (FP32) operations and improved power efficiency provide significant performance improvements for graphics and simulation workflows, such as complex 3D computer-aided design (CAD) and computer-aided engineering (CAE).
Second-Generation RT Cores
With up to 2x the throughput over the previous generation and the ability to concurrently run ray tracing with shading or denoising, second-generation RT cores deliver massive speedups for workloads like photorealistic rendering of movie or television content, architectural design evaluations, and virtual prototyping of product designs. This technology also speeds up the rendering of ray-traced motion blur for faster results with greater visual accuracy.
Third-Generation Tensor Cores
New Tensor Float 32 (TF32) precision provides up to 5x the training throughput over the previous generation to accelerate AI and data science model training without requiring any code changes. Hardware support for structural sparsity provides up to 10x throughput for inferencing. Tensor Cores also accelerate AI augmented graphics capabilities like DLSS (Deep learning Super Sampling), AI denoising, and enhanced editing for select applications.
48 GB of GPU Memory
Ultra-fast GDDR6 memory, scalable up to 96 GB with NVLink, gives data scientists, engineers, and creative professionals the large memory necessary to work with massive datasets and workloads like ultra-high resolution rendering, data science and simulation.
Third-Generation NVIDIA NVLink
Connect two A40 GPUs together to scale from 48 GB of GPU memory to 96 GB. Increased GPU-to-GPU interconnect bandwidth (up to 112.5 GB/s) provides a single scalable memory to accelerate graphics and compute workloads and tackle larger datasets. A new, more compact NVLink connector enables deployment in a wider range of servers.
Next generation improvements with NVIDIA virtual GPU (vGPU) software, such as the NVIDIA Virtual Workstation, allows for larger, more powerful virtual workstation instances for remote users, enabling high-end remote design, AI, and compute workloads.
PCI Express Gen 4
PCI Express Gen 4 doubles the bandwidth of PCIe Gen 3, improving data-transfer speeds from CPU memory for data-intensive tasks like AI, data science, and 3D design. Faster PCIe performance also accelerates GPU direct memory access (DMA) transfers, providing faster I/O communication of video data between the GPU and GPUDirect for Video enabled devices, delivering a powerful solution for live broadcast. A40 is backwards compatible with PCIe Express Gen 3 for deployment flexibility.
Data Center Efficiency and Security
Featuring a dual-slot, power efficient design, NVIDIA A40 is up to 2x as power efficient as the previous generation that is validated with a wide range of NVIDIA-certified systems from PNY backed systems builders and integrators. The NVIDIA A40 also provides secure and measured boot with hardware root of trust capability, ensuring that firmware has not been tampered with or corrupted.
NVIDIA A40 Specifications:
|48GB GDDR6 with ECC (Error Correcting Code)
|PCIe Gen 4 x16 | 31.5 GB/s Bidirectional
|2-way Low Profile (2-Slot) | 112.5 GB/s Bidirectional
|3x DisplayPort 1.4*
|Max Power Consumption
|4.4” H x 10.5” L | Dual Slot
|vGPU Software Support
|NVIDIA vPC/vApps, NVIDIA RTX Virtual Workstation, NVIDIA Virtual Compute Server, NVIDIA AI Enterprise
|vGPU Profiles Supported
|See the Virtual GPU Licensing Guide
|NVEC | NVDEC
|1x | 2x (includes AV1 decode)
|Secure and Measured Boot
|Yes | With Hardware Root of Trust
* A40 is configured for virtualization by default with physical display connectors disabled. The display outputs can be enabled via management software tools.
NVIDIA A40 combines the performance and features necessary for large-scale display experiences, VR, broadcast-grade streaming, and more.
Drive massive CAVE Automatic virtual Environments (CAVEs), video walls, virtual sets and broadcast, and location-based entertainment deployments with support for multiple 8K monitors, NVIDIA Mosaic multi-display technology with bezel correction, and NVIDIA’s Warp and Blend SDK.
NVIDIA Quadro Sync II lets you synchronize multiple NVIDIA A40 GPUs with displays or projectors to create large-scale visualization environments, Frame Lock GPU outputs for artifact free images across multiple displays, or synchronize GPU outputs to an external timing source.
Video Encode and Decode
With dedicated video encoder (NVENC) and decoder engines (NVDEC), access the performance needed to work with multiple streams simultaneously, export video faster, and use multi-stream applications for broadcast, security, and video serving.
Power the most immersive Augmented Reality (AR) and Virtual Reality (VR) experiences on the highest-resolution Head-Mounted Displays (HMDs) with accelerated graphics and increased display bandwidth. Four-way VR SLI enables peak performance, assigning 2 NVLink connected A40 GPUs to each eye.
Enterprise Drivers and ISV Certifications
Virtual workstations powered by NVIDIA RTX Virtual Workstation (vWS) software leverage the same NVIDIA RTX platform as physical workstations, benefiting from extensive tenting across a broad range of industry applications and certifications from over 100 Independent Software Vendor (ISV) applications to ensure optimal performance and stability.
EGX Platform for Professional Visualization
The NVIDIA A40, along with NVIDIA vGPU software, is at the heart of the next-generation NVIDIA EGX Platform, which is certified by NVIDIA for professional visualization, delivering the performance and features that can power professional graphics and computing anywhere.
NVIDIA A40 is designed, built, and tested by NVIDIA for the most demanding data center visualization, AI, simulation, and compute workloads, and is now available from PNY Technologies. PNY’s willingness to stock the A40 cuts lead times, and PNY’s world class pre- and post-sales support will ensure your deployment(s) to work. Contact PNY at email@example.com for additional information.
Make sure you register for NVIDIA GTC 21. This industry renowned event connects you with experts from NVIDIA to learn about the latest breakthroughs in AI, data center, accelerated computing, healthcare, intelligent networking, virtual reality, applications development and more. You can't miss the GTC keynote, delivered by Jensen Huang, NVIDIA Founder, President, and CEO, so save the date now!