NVIDIA ConnectX-7 NDR200 200G Dual-Port QSFP112 InfiniBand Adapter
Dual-port 200G QSFP112 NIC balancing bandwidth, resiliency, and density.
We help you choose, configure, and deliver the right system — no obligation.
Configuration at a Glance
Tailored per engagement. Full technical overview below.
Configuration Options
Core specifications for this system. Every component is configurable to your workload — request a quote for a tailored build.
RDMA, GPUDirect RDMA and Storage
Overview
The ConnectX-7 NDR200 is a dual-port QSFP112 adapter delivering 200Gb/s per port over PCIe Gen5 x16, supporting both NDR200 InfiniBand and 200GbE for flexible cluster connectivity. Nexus Compute specifies, configures, and validates each card against your platform and protocol mode, delivered warranty-backed through NVIDIA-authorized channels.
Who This Solution Is For
Business Benefits
Resilient dual-port design
Two independent 200G ports provide path redundancy and aggregated bandwidth for production nodes.
Protocol flexibility
A single card serves NDR200 InfiniBand or 200GbE, simplifying standardization across mixed fabrics.
Validated to your stack
Each adapter is configured and tested for your server, mode, and firmware before warranty-backed delivery.
Typical Business Use Cases
Dual-rail NDR200 leaf connectivity
Inference and fine-tuning server fabric
RoCE 200GbE storage networking
Redundant host connectivity for HA nodes
Industry Applications
Technical Overview
The ConnectX-7 NDR200 adapter provides two QSFP112 ports at 200Gb/s each over a PCIe Gen5 x16 host interface, switchable between NDR200 InfiniBand and 200GbE with secure boot enabled. In-network computing, RDMA, and GPUDirect support deliver low-latency data movement for both training-adjacent and inference workloads.
| Network Ports | 2x 200Gb/s NDR200 / 200GbE (QSFP112) |
| Host Interface | PCIe 5.0 x16 (x16 extension option) |
| Protocols | InfiniBand NDR200 and Ethernet (VPI) |
| RDMA / GPUDirect | RDMA, GPUDirect RDMA and Storage |
| Form Factor | PCIe HHHL, tall/short bracket |
| Security | Secure boot enabled, crypto-disabled SKU |
| Acceleration | In-network computing, MTU up to 4KB |
| Warranty | Manufacturer-backed via authorized channels |
Specifications are indicative and configured to each engagement. Request a quote for a configuration tailored to your requirements.
Warranty, Support & Fulfillment
Every system ships from an authorized channel, configured and tested, with the documentation enterprise buyers need — backed by warranty and a dedicated account team.
Enterprise Warranty
Full manufacturer warranty with optional on-site, next-business-day support and extended coverage.
Authorized Channel
Sourced through Tier-1 distribution and OEM partners — never grey market. Asset & warranty records included.
Lead Time & Deployment
48-hour quotes, then configured, burn-in tested, and delivered on a committed schedule.
Nationwide Fulfillment
Coordinated logistics, rack-and-stack, and delivery wherever your infrastructure lives.
Frequently Asked Questions
When is NDR200 the right choice over full NDR 400G?
NDR200 suits inference, fine-tuning, and storage tiers where 200G per port is ample and dual-port redundancy matters more than peak per-GPU rail bandwidth.
Can this adapter run Ethernet as well as InfiniBand?
Yes, it is a VPI card that operates in NDR200 InfiniBand or 200GbE mode; we preconfigure the port mode to match your fabric.
Does the dual-port design double effective bandwidth?
The two ports can aggregate for up to 400Gb/s of host bandwidth or run split for redundancy; the achievable rate depends on PCIe Gen5 x16 lane budget, which we verify per server.
Hardware Assistance
Configure the NVIDIA ConnectX-7 NDR200 200G Dual-Port QSFP112 InfiniBand Adapter with Nexus Compute
Tell us your requirements and a hardware specialist will help you specify, configure, and quote the right system — typically within two business days. No obligation.