Grifo Model Vector A10090
The best on the market
to accelerate Artificial Intelligence (AI), High Performance Computing (HPC), Data Science and Graphics.
Making real what is impossible with any existing system.
GRIFO Model Vector A10090 massive compute system in numbers
GRIFO Model Vector A10090 integrates 90 NVIDIA V100s Tensor Core GPUs, and 3.6 Terabyte of GPU memory so you can work with the biggest training datasets.
Multiple balanced NVMe array
GRIFO Model Vector A10090 integrates 72 NVMe PCIe 4x enabling parallel data access at the speed of more than 200 Gyte/s sustained.
Ultra fast data exchange
GRIFO Model Vector A10090 delivers up to 144 TeraBytes GPU memory bandwidth.
Every time you have a real bottleneck to solve, you have to find a new architecture.