Exxact Corporation Extends Its Line of AI Development Servers to Support Latest NVIDIA a100 PCIe GPU
Exxact Corporation, a leading provider of high-performance computing (HPC), artificial intelligence (AI), and data center solutions, announced that they are now offering a wide selection of servers for AI featuring the new NVIDIA A100 PCIe GPU. This new line of servers will allow researchers, developers, enterprises, and university labs to upgrade existing infrastructure or quickly deploy a state-of-the-art deep learning platform.
The NVIDIA A100 Tensor Core GPU is at the heart of the new server line, so users can tackle tough computing challenges and deliver unprecedented acceleration at every scale for AI, data analytics, and HPC. The A100 GPU can scale up to thousands of GPUs or, using new Multi-Instance GPU (MIG) technology, can be partitioned into up to seven isolated GPU instances to accelerate workloads of all sizes.
Recommended AI News: Facet Digital Transforms Business Operations With Couchbase Cloud
The first GPU based on the NVIDIA Ampere architecture, the A100 can boost compute performance by up to 20x over its predecessor — making it the company’s largest leap in GPU performance to date. It features several technical breakthroughs, including the aforementioned MIG technology for varying sized compute jobs, and new structural sparsity capabilities that can be used to double a GPU’s compute performance. Furthermore, the NVIDIA A100 supports PCI Express Gen 4 (PCIe Gen 4) which doubles the bandwidth of PCIe 3.0/3.1.
“With the release of the A100, NVIDIA has again set the standard for GPU performance, scalability, and efficiency,” said Jason Chen, President of Exxact Corporation. “Our AI training infrastructure comes pre-loaded with popular deep learning frameworks via our EMLI (Exxact Machine Learning Images) software stack, which allows our customers to realize massive performance gains of the A100 right out of the box.” In addition, users also have access to the latest versions of GPU-optimized AI, machine learning, and HPC software from NVIDIA NGC, allowing users to build their solutions faster and achieve breakthroughs sooner.
Exxact Corporation will offer both Intel Xeon and AMD EPYC-based server options including the following:
NVIDIA A100 PCIe Servers Featuring AMD EPYC Processors
- 4x NVIDIA GPU Server with Single AMD EPYC Processor
- 4x NVIDIA GPU Server with Dual AMD EPYC Processors
- 8x NVIDIA GPU Server with Dual AMD EPYC Processors
NVIDIA A100 PCIe Servers Featuring Intel Xeon Scalable Processors
- 4x NVIDIA GPU Server with Dual Intel Xeon Processors
- 8x NVIDIA GPU Server with Dual Intel Xeon Processors
- 10x NVIDIA GPU Server with Dual Intel Xeon Processors
Recommended AI News: Daily AI Roundup: The 5 Coolest Things On Earth Today