Unlock Ultimate Computational Power, pre-reserve the all new NVIDIA H200s AI SuperCluster for $1.99/hr. Click to Know More.
InfiniBand
InfiniBand is a technology that facilitates the seamless connection of servers
equipped with GPUs within a cluster. Use GPU clusters with InfiniBand interconnect to
accelerate your data-intensive ML workloads.
Get a multi-node infrastructure with up to 3.2 Tbit/s of per-host networking performance.
InfiniBand provides the ultra-high data transfer speeds essential for ML workflows that require rapid processing and analysis of large volumes of data.
Large-scale clustering and parallel processing make InfiniBand indispensable for complex and resource-intensive algorithms.
High reliability and fault tolerance provided by InfiniBand ensure that ML workflows run smoothly with minimal downtime.
InfiniBand offers low latency communication required by real-time ML applications with quick decision-making and response times.
InfiniBand can be a cost-effective solution for engineers who need high-performance networking capabilities without investing in expensive proprietary hardware.
We provide GPU clusters with NVIDIA® Hopper® H100 SXM GPUs.
A host consists of 8 GPUs. Each GPU features up to 400 Gbit/s connection, providing up to 3.2 Tbit/s network bandwidth per host.
You can use Marketplace images adapted for GPU clusters, e.g. Ubuntu 22.04 LTS for NVIDIA® GPU clusters (CUDA® 12).
