Nvidia DGX
NVIDIA DGX A100 rental
The NVIDIA DGX A100 is a server specially designed for AI workloads that require high performance. The system delivers an unprecedented density and is in terms of hardware and software based on the supercomputer of NVIDIA, the DGX SATURNV. The DGX A100 is equipped with 8x NVIDIA A100 Tensore GPUs. The most advanced GPU, available in a 40GB and 80GB variant. The A100 offers more compute, more FLOPS and extremely high performance. This packed in a compact 6U chassis, with which enterprises can now implement the AI infrastructure even easier.
DGX A100 Specs
- A100 GPU offers more on-chip memory, including a 40MB Level 2 cache;
- DGX A100 introduces next generation NVIDIA NVLink;
- Next generation NVLink doubles GPU-to-GPU direct bandwidth to 600(GB/s) gigabytes per second (10X higher than PCIe Gen 4);
- TF32 works like FP32 and provides 20X more FLOPS for AI compared to the previous generation (no code changes are required to achieve this speed increase);
- A100 GPU provides 1.6 (TB/s) terabytes per second of memory bandwidth.
NVIDIA DGX Station A100 rental
The NVIDIA DGX Station A100 is a server station specially designed for AI workloads that require high performance. The system delivers an unprecedented density and is in terms of hardware and software based on the supercomputer of NVIDIA, the DGX SATURNV. The DGX Station A100 is equipped with 4x NVIDIA A100 Tensore GPUs. The most advanced GPU, available in a 40GB and 80GB variant. The A100 offers more compute, more FLOPS and extremely high performance. This packed into a compact 6U chassis, enterprises can now implement AI infrastructure even easier.
Specs of the DGX Station A100
- AI workgroup server with 2.5 petaFLOPS of performance
- World-class AI platform with no complicated installation or IT support needed\
- With four full NVIDIA A100 Tensor Core GPUs and up to 320 gigabytes (GB) of GPU memory