Limited Time Offer: Save $350 on your Ai Cluster → Shop Now

--

Days

--

Hours

--

Mins

--

Secs

Sale Ended

TECHNOLOGY AT A GLANCE

Explore the capabilities

2 × 200G QSFP112 ports

Two QSFP112 ports provide up to 400Gb/s of aggregate network bandwidth.

NDR200 InfiniBand

Support compatible NDR200 InfiniBand environments for demanding compute and storage clusters.

PCIe Gen4 and Gen5

A PCIe x16 host interface provides high-speed connectivity to compatible servers.

RDMA acceleration

RDMA and GPUDirect RDMA help move data efficiently in supported GPU, storage, and compute workflows.

Dual-port 200G connectivity for high-performance infrastructure

Connect two 200G network links to compatible InfiniBand or Ethernet fabrics for high-bandwidth AI, HPC, storage, and cloud deployments.

TECHNICAL DETAILS

Built into the system

Dual-port 200G connectivity

Two QSFP112 ports support up to 200Gb/s per port, providing up to 400Gb/s of aggregate network bandwidth for compatible high-performance infrastructure.

InfiniBand and Ethernet flexibility

The adapter supports compatible NDR200 InfiniBand and Ethernet deployments, allowing it to fit into specialized compute fabrics or high-speed Ethernet networks.

PCIe 4.0 and PCIe 5.0 x16 host interface

A PCIe x16 interface supports installation in compatible PCIe Gen4 or Gen5 server platforms for high-bandwidth host connectivity.

RDMA and GPU-aware data movement

RDMA and GPUDirect RDMA help reduce CPU involvement in supported AI, HPC, storage, and GPU-accelerated workloads.

Software-defined networking acceleration

ASAP technology supports hardware acceleration for compatible virtual switching and software-defined networking environments.

Traffic control and reliability

End-to-end QoS and congestion-control capabilities help maintain efficient traffic flow across compatible high-bandwidth networks.