@Xi Logo @Xi Logo
Navigation Navigation
List of Xi® Servers available with NVIDIA® A100 powered by Ampere™
A100 image
The Most Powerful End-to-End AI and HPC Data Center Platform

The NVIDIA A100 Tensor Core GPU delivers unprecedented acceleration at every scale for AI, data analytics, and high-performance computing (HPC) to tackle the world’s toughest computing challenges. As the engine of the NVIDIA data center platform, A100 can efficiently scale to thousands of GPUs or, with NVIDIA Multi-Instance GPU (MIG) technology, be partitioned into seven GPU instances to accelerate workloads of all sizes. And third-generation Tensor Cores accelerate every precision for diverse workloads, speeding time to insight and time to market.
Deep Learning Training

NVIDIA A100’s third-generation Tensor Cores with Tensor Float (TF32) precision provide up to 20X higher performance over the prior generation with zero code changes and an additional 2X boost with automatic mixed precision and FP16. When combined with third-generation NVIDIA® NVLink®, NVIDIA NVSwitch™, PCI Gen4, NVIDIA Mellanox InfiniBand, and the NVIDIA Magnum IO™ software SDK, it’s possible to scale to thousands of A100 GPUs. This means that large AI models like BERT can be trained in just 37 minutes on a cluster of 1,024 A100s, offering unprecedented performance and scalability..

High-Performance Computing

A100 introduces double-precision Tensor Cores, providing the biggest milestone since the introduction of double-precision computing in GPUs for HPC. This enables researchers to reduce a 10-hour, double-precision simulation running on NVIDIA V100 Tensor Core GPUs to just four hours on A100. HPC applications can also leverage TF32 precision in A100’s Tensor Cores to achieve up to 10X higher throughput for single-precision dense matrix multiply operations.

Enterprise-Ready Utilization

A100 with MIG maximizes the utilization of GPU-accelerated infrastructure like never before. MIG allows an A100 GPU to be partitioned into as many as seven independent instances, giving multiple users access to GPU acceleration for their applications and development projects. MIG works with Kubernetes, containers, and hypervisor-based server virtualization with NVIDIA Virtual Compute Server (vComputeServer). MIG lets infrastructure managers offer a right-sized GPU with guaranteed quality of service (QoS) for every job, optimizing utilization and extending the reach of accelerated computing resources to every user.

To learn more about the NVIDIA A100 Tensor Core GPU, visit www.nvidia.com/a100
© 2020 NVIDIA Corporation. All rights reserved. NVIDIA, the NVIDIA logo, CUDA, DGX, HGX, HGX A100, NVLink, NVSwitch, OpenACC, TensorRT, and Volta are trademarks and/or registered trademarks of NVIDIA Corporation in the U.S. and other countries. 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. Jun20
Model name   NVIDIA A100 for PCIe
Peak FP64   9.7 TF
Peak FP64 Tensor Core   19.5 TF
Peak FP32   19.5 TF
Peak TF32 Tensor Core   156 TF | 312 TF*
Peak BFLOAT16 Tensor Core   312 TF | 624 TF*
Peak FP16 Tensor Core   312 TF | 624 TF*
Peak INT8 Tensor Core   624 TOPS | 1,248 TOPS*
Peak INT4 Tensor Core   1,248 TOPS | 2,496 TOPS*
GPU Memory   40 GB
GPU Memory Bandwidth   1,555 GB/s
Interconnect   NVIDIA NVLink 600 GB/s PCIe Gen4 64 GB/s **
Multi-instance GPUs   Various instance sizes with up to 7MIGs @5GB
Form Factor   PCIe
Max TDP Power   250W
Delivered Performance of Top Apps   90%
* With sparsity
 ** PCIe GPUs via NVLink Bridge for up to 2-GPUs:

Ampere family
NVIDIA A100 for PCIe - 900-21001-0000-000
Best Optimized Data Center Servers for NVIDIA A100 for PCIe GPU
Xi® NetRAIDer™ 64LT
Xi® NetRAIDer™ 64XLT
Xi® NetRAIDer™ 64XE
Xi® WebRAIDer™ 64X-1U

*Published Prices. Prices shown (provided by way of a Quotation or a Pre-Configured List) are subject to change without prior notice to the prices in effect at the time of Quotation or placing an Order. Seller reserves the right to make any corrections to prices quoted due to availability, clerical errors or errors of omission. In the event of any specific requirements (including without limitation any design, specification, ordered quantity, or shipment changes) representing a price increase.
Follow us on @Xi Social Media:
Facebook Logo Twitter Logo Linkedin Logo Instagram Logo Youtube Logo Sign up for updates | Request more info TAA LogoGSA LogoIntel TPPNPN Logo .:Total views:.
All trademarks and brands mentioned on this website may be legally registered in the U.S. and/or other countries. They are subject without restriction to the terms of applicable registered trademark rights and the ownership rights of the respective registered owners. The mention of a trademark should not be taken to indicate that such a trademark is not subject to third-party rights.
Home | Products | Support | Company | Contact | Terms & Privacy © 1996- @Xi® Computer Corporation | All Rights reserved.