Equipped with four NVIDIA GPU V100s cards and 32GB of dedicated memory, this platform offers exceptional performance to accelerate data analytics and artificial intelligence (AI) workflows. It is suitable for training and inference calculation, high performance computing (HPC) and image rendering.
It also has flexible memory and storage options, so you can adapt your server to suit your needs. Dual power supplies, a redundant 50Gbps high-speed network, and a 99.99% SLA ensure that your infrastructure has a high level of efficiency and reliability.
Performance (per GPU card): double precision: 8.2Tflops, single precision: 16.4Tflops, and in deep learning: 130Tflops. Memory bandwidth: 1,134Gbps.
- Dedicated system disks
- Hot-swap technology
- OVHcloud Link Aggregation (OLA)
AI, Machine Learning & Deep learning
High Performance Computing (HPC)