Complex workloads demand ultra-fast processing of high-resolution simulations, extreme-size datasets, and highly parallelized algorithms. As these computing requirements continue to grow, NVIDIA Quantum InfiniBand—the world’s only fully offloadable, In-Network Computing platform—provides the dramatic leap in performance needed to achieve unmatched performance in high-performance computing (HPC), AI, and hyperscale cloud infrastructures with less cost and complexity.
InfiniBand Networking Solutions
InfiniBand Adapters
InfiniBand host channel adapters (HCAs) provide ultra-low latency, extreme throughput,
and innovative NVIDIA In-Network Computing engines to deliver the acceleration, scalability, and feature-rich technology needed for today's modern workloads.
Data Processing Units (DPUs)
The NVIDIA® BlueField® DPU combines powerful computing, high-speed networking,
and extensive programmability to deliver software-defined, hardware-accelerated solutions for the most demanding workloads. From accelerated AI computing to cloud-native supercomputing, BlueField redefines what’s possible.
InfiniBand Switches
InfiniBand switch systems deliver the highest performance and port density available. Innovative capabilities such as NVIDIA Scalable Hierarchical Aggregation and Reduction Protocol (SHARP)™ and advanced management features such as self-healing network capabilities, quality of service, enhanced virtual lane mapping, and NVIDIA In-Network Computing acceleration engines provide a performance boost for industrial, AI, and scientific applications.
Routers and Gateway Systems
InfiniBand systems provide the highest scalability and subnet isolation using InfiniBand routers, and InfiniBand to Ethernet gateway systems. The latter is used to enable a scalable and efficient way to connect InfiniBand data centers to Ethernet infrastructures.
Long-Haul Systems
NVIDIA MetroX® long-haul systems can seamlessly connect remote InfiniBand data centers, storage, and other InfiniBand platforms. They can extend the reach of InfiniBand up to 40 kilometers, enabling native InfiniBand connectivity between remote data centers
or between data center and remote storage infrastructures for high availability and disaster recovery.
Cables and Transceivers
LinkX® cables and transceivers are designed to maximize the performance of HPC networks, requiring high-bandwidth, low-latency, highly reliable connections between InfiniBand elements.
InfiniBand-Enhanced Capabilities
In-Network Computing
NVIDIA Scalable Hierarchical Aggregation and Reduction Protocol (SHARP)™ offloads collective communication operations to the switch network, decreasing the amount of data traversing the network, reducing the time of Message Passing Interface (MPI) operations, and increasing data center efficiency.
Self-Healing Network
NVIDIA InfiniBand with self-healing network capabilities overcomes link failures, enabling network recovery 5,000X faster than any other software-based solution. These capabilities take advantage of the intelligence built into the latest generation of InfiniBand switches.
Quality of Service
InfiniBand is the only high-performance interconnect solution with proven quality-of-service capabilities, including advanced congestion control and adaptive routing, resulting in unmatched network efficiency.
Network Topologies
InfiniBand offers centralized management and supports any topology, including Fat Tree, Hypercubes, multi-dimensional Torus, and Dragonfly+. Routing algorithms optimize performance when designing a topology for particular application communication patterns.
Resources
-
Solution Briefs
-
Videos