Click to Enlarge
NVIDIA Networking Platforms
NVIDIA’s end-to-end accelerated networking platforms offer organizations the choice of NVIDIA Quantum InfiniBand and Spectrum™ Ethernet connectivity to provide the highest performance for AI training and inference workloads. Innovations such as adaptive routing and telemetry-based congestion control enable consistent, predictable performance with zero tail latency for AI workloads scaling to 100K+ GPUs.
Accelerated Networking Platforms
Quantum InfiniBand Platform
Highest networking performance and scale up to 800Gb/s
Develop solutions with in-network computing, custom topologies, and self-healing fabrics
Spectrum Ethernet Platform
Accelerated, high-performance Ethernet up to 800Gb/s
Leverage Spectrum-X as the world’s first Ethernet platform built for AI
Enable software-driven automation, simulation, and validation for operational efficiency
BlueField Platform
Leverage smart offloads for cloud, security, timing, video, and storage
Accelerate applications at speeds up to 400 Gb/s
Accelerated Networking Software and Tools
NVIDIA DOCA
NVIDIA DOCA™ unlocks the potential of the NVIDIA® BlueField® networking platform. DOCA enables the rapid creation of applications and services that offload, accelerate, and isolate data center workloads.
NVIDIA Rivermax
NVIDIA® Rivermax® offers a unique IP-based solution for any media and data streaming use case. Rivermax unlocks innovation for a wide range of applications in media and entertainment (M&E), broadcast, healthcare, smart cities, and more.
NVIDIA HPC-X
NVIDIA® HPC-X® is a comprehensive software package that enables MPI and SHMEM/PGAS programming languages to achieve high performance, scalability, and efficiency. HPC-X also ensures that communication libraries are fully optimized by NVIDIA Quantum InfiniBand networking solutions.
NVIDIA Magnum IO SDK
The NVIDIA Magnum IO™ software development kit (SDK) enables developers to remove input/output (IO) bottlenecks in AI, high-performance computing (HPC), data science, and visualization applications, reducing the end-to-end time of their workflows.