InfiniBand Enables the Most Efficient Machine Learning Platforms
Machine learning is a pillar of today’s technological world, offering solutions that enable better
and more accurate decision-making based on the great amounts of data being collected. Machine
learning encompasses a wide range of applications, ranging from security, finance, and image and
voice recognition, to self-driving cars, healthcare and smart cities.
InfiniBand accelerates all popular frameworks such as TensorFlow, CNTK, Paddle, Pytorch
and Apache Spark with RDMA, and continues to innovate and accelerate solutions for fastest and most
scalable distributed execution of training large and powerful models.
By providing low latency, high bandwidth, high message rate, and smart offloads, InfiniBand solutions are the most deployed high-speed interconnect for large-scale machine learning
- for both training and inferencing systems.
The Selene Supercomputer
NVIDIA’s DGX SuperPOD with NVIDIA Mellanox HDR 200Gb/s InfiniBand Deployment
One of the fastest and most efficient supercomputer on the planet
built in under one month
Maximizing Data Center Storage and Network IO Performance with NVIDIA Magnum IO
Magnum IO utilizes storage IO, network IO, in-network compute, and IO management to simplify and speed up data movement, access, and management for multi-GPU, multi-node systems.
Magnum IO supports NVIDIA CUDA-X™ libraries and makes the best use of a range of NVIDIA GPU and NVIDIA networking hardware topologies to achieve optimal throughput and low latency.
Learn More
SETTING A NEW BAR IN MLPERF
NVIDIA training and inference solutions deliver record-setting performance in MLPerf, the leading industry benchmark for AI performance.
NVIDIA RAPIDS Accelerator and InfiniBand Supercharges Apache Spark 3.0

As data scientists shift from using traditional analytics to AI applications with Apache Spark 3.0, InfiniBand provides the fastest path to move data between remote Spark nodes and is tightly coupled to the Spark Shuffle operations leveraging the UCX communication framework.
Learn More10X Performance with NVIDIA Mellanox GPUDirect® RDMA
Designed specifically for the needs of GPU acceleration, GPUDirect RDMA provides direct communication between NVIDIA GPUs in remote systems. This eliminates the system CPUs and required buffer copies of data via the system memory, resulting in 10X better performance.

- NVIDIA Mellanox Product DocumentationView
- NVIDIA Mellanox UFM PlatformDownload
- NVIDIA Mellanox MetroX-2 SystemsDownload
- QM8700 Mellanox Quantum™ HDR Edge SwitchDownload
- QM8790 Mellanox Quantum™ HDR Edge SwitchDownload
- CS8500 Mellanox Quantum™ HDR Modular SwitchDownload
- ConnectX-6 HDR 200Gb/s VPI AdapterDownload
- LinkX® InfiniBand Cables and TransceiversView
- BlueField-2Download
- BlueFieldDownload
- Innova SmartNICDownload