 
|
NVIDIA Mellanox MQM9790-NS2F 64-Port NDR 400Gb/s InfiniBand Router | 51.2Tb/s Throughput with Advanced Routing & P2C Airflow
Product Details:
| Brand Name: | Mellanox |
| Model Number: | MQM9790-NS2F(920-9B210-00FN-0D0) |
| Document: | MQM9700 series.pdf |
Payment & Shipping Terms:
| Minimum Order Quantity: | 1pcs |
|---|---|
| Price: | Negotiate |
| Packaging Details: | outer box |
| Delivery Time: | Based on inventory |
| Payment Terms: | T/T |
| Supply Ability: | Supply by project/batch |
|
Detail Information |
|||
| Application: | Server | Condition: | New And Original |
|---|---|---|---|
| Aggregate Bandwidth: | 51.2 Tb/s Bidirectional | Management: | MLNX-OS With CLI, WebUI, SNMP, JSON Support |
| Name: | MQM9790-NS2F InfiniBand Mellanox Network Switch Scaling Out Data Centers With NDR 400G InfiniBand Smart Switches | Keyword: | Mellanox 400g Switch |
| Highlight: | NVIDIA Mellanox 400Gb/s InfiniBand router,64-port NDR InfiniBand switch,51.2Tb/s throughput network switch |
||
Product Description
NVIDIA Mellanox MQM9790-NS2F 64-Port NDR 400Gb/s InfiniBand Router
High-performance 64-port Quantum-2 managed router with advanced routing and front-to-rear cooling for large-scale HPC and AI fabrics
Product Overview
The NVIDIA Mellanox MQM9790-NS2F is a high-performance InfiniBand router delivering 64 ports of 400Gb/s full bidirectional bandwidth in a compact 1U form factor. Featuring P2C (front-to-rear) airflow and advanced routing capabilities, this model is purpose-built for large-scale HPC, AI, and multi-tenant data center environments requiring network segmentation, isolation, and inter-subnet routing.
As a managed router in the NVIDIA Quantum-2 series, it provides 64 QSFP56 ports running at 400Gb/s per port with advanced routing functionality including InfiniBand subnet routing, traffic isolation, and multi-tenant network segmentation. The integrated x86 dual-core CPU runs MLNX-OS, enabling comprehensive on-box management including integrated subnet manager and advanced routing protocols for large-scale fabrics.
Key Features
- 64 x 400Gb/s Ports: Full bidirectional bandwidth per port with QSFP56 connectors, supporting passive copper, active optical cables, and optical modules
- 51.2Tb/s Throughput: Non-blocking aggregate switching capacity with ultra-low port-to-port latency
- Advanced InfiniBand Routing: Supports subnet routing, traffic isolation, and multi-tenant network segmentation for complex fabric architectures
- In-Network Computing (SHARP v3): Hardware-accelerated collective offload reduces MPI operations time and improves HPC application scalability
- Managed with MLNX-OS: On-board dual-core x86 CPU with 8GB RAM and 32GB SSD; full chassis management via CLI, SNMP, and optional UFM integration
- 1+1 Redundant Power & Fans: Hot-swappable 80 Gold+ and ENERGY STAR certified power supplies; P2C (front-to-rear) airflow with hot-swappable fans
- High Efficiency: Optimized power design with ATIS weighted power consumption as low as 180W for a fully populated system
- Large-Scale Fabric Support: Designed for fabrics with thousands of nodes, providing robust routing and forwarding capabilities
Router Capabilities
The MQM9790-NS2F provides advanced InfiniBand routing features beyond standard switching:
- Subnet-to-subnet routing for multi-tier fabrics
- Traffic isolation and virtual fabric support
- Multi-tenant network segmentation
- Enhanced forwarding tables for large-scale deployments
- Integration with UFM for routing policy management
Airflow Direction: P2C (Front-to-Rear)
This model draws cool air from the front of the chassis and exhausts hot air through the rear. It is intended for standard data center environments with front cold-aisle configurations. Please ensure your rack cooling configuration matches this airflow direction. For rear-to-front cooling, order MQM9790-NS2R.
Technology: In-Network Computing with SHARP v3
The NVIDIA SHARP v3 technology embedded in the Quantum-2 ASIC enables the MQM9790-NS2F to perform collective operations such as MPI reductions directly within the network fabric. This eliminates multiple data traversals between endpoints, drastically reducing MPI operation time and freeing CPU cores for application computation. For fully populated fat-tree networks, partial reduction operations can complete in under three microseconds.
Typical Deployments
- Large-Scale HPC Clusters: Interconnecting multiple subnets in petascale and exascale computing environments
- Multi-Tenant AI Infrastructure: Providing network isolation and segmentation for AI training and inference workloads across multiple teams
- Spine Routing in Fat-Tree Fabrics: Serving as a routing layer between leaf switches in large-scale InfiniBand networks
- Enterprise Data Centers: Supports high-performance storage, database, and compute workloads requiring low latency and high throughput
- Cloud and Bare-Metal Infrastructures: Ideal for multi-tenant environments requiring high-bandwidth, low-latency network fabrics with tenant isolation
Key Performance Facts
64 x NDR 400Gb/s
Ports
51.2Tb/s
Throughput
Under 90ns
Latency
Router
Type
P2C (Front-to-Rear)
Airflow
On-box (MLNX-OS)
Management
180W - 200W
Power (ATIS)
1U, 27" depth
Form Factor
Technical Specifications
| Parameter | Specification |
|---|---|
| Model | MQM9790-NS2F |
| Series | NVIDIA Quantum-2 |
| Product Type | InfiniBand Router |
| Ports | 64 x QSFP56 (NDR 400Gb/s) |
| Aggregate Throughput | 51.2Tb/s (non-blocking) |
| Latency | Under 90ns (port-to-port) |
| Switch ASIC | NVIDIA Quantum-2 NDR |
| Routing Capability | Advanced InfiniBand routing with multi-subnet support |
| CPU | Intel x86 dual-core |
| System Memory | 8GB RAM, 32GB SSD |
| Management | MLNX-OS (on-box), CLI, SNMP, UFM |
| Power Supplies | 1+1 redundant, hot-swappable, 100-240VAC, 80 Gold+ |
| Typical Power (ATIS) | 180W - 200W (fully populated) |
| Cooling | P2C (front-to-rear), hot-swappable fans |
| Form Factor | 1U rackmount; 43.6mm (H) x 433.2mm (W) x 590.6mm (D) |
| Weight | 12.5 kg (approx.) |
| Operating Temperature | 0°C to 45°C |
| Certifications | CE, FCC, VCCI, ICES, RCM, RoHS, CB, cTUVus, CU |
Compatibility Matrix
| Component | Compatible Products |
|---|---|
| Adapters | NVIDIA ConnectX-6 / ConnectX-7 (NDR, HDR, EDR, FDR) |
| Cables | QSFP56 DAC, AOC, optical transceivers (NVIDIA LinkX) |
| Management | NVIDIA UFM, MLNX-OS, SNMP v1/v2/v3, CLI |
| Standards | IBTA 1.4, 1.3; RoHS-6; ENERGY STAR |
| Routing | InfiniBand subnet routing, multi-tenant isolation |
Selection Guide
| Ordering Part Number | Description | Airflow | Type |
|---|---|---|---|
| MQM9700-NS2F | 400Gb/s InfiniBand 1U switch, 64 QSFP56, x86 dual-core | P2C (front-to-rear) | Switch |
| MQM9700-NS2R | 400Gb/s InfiniBand 1U switch, 64 QSFP56, x86 dual-core | C2P (rear-to-front) | Switch |
| MQM9790-NS2F | 400Gb/s InfiniBand 1U router, 64 QSFP56, x86 dual-core | P2C (front-to-rear) | Router |
| MQM9790-NS2R | 400Gb/s InfiniBand 1U router, 64 QSFP56, x86 dual-core | C2P (rear-to-front) | Router |
Note: The MQM9790 series includes advanced routing capabilities. Verify airflow direction (P2C or C2P) matches your data center hot/cold aisle configuration before ordering.
Key Advantages
- Industry-leading performance: 51.2Tb/s throughput and ultra-low latency in a compact 1U chassis
- Advanced routing capabilities: Enables multi-subnet fabrics, tenant isolation, and large-scale network segmentation
- Reduced IT costs: High port density (64 ports in 1U) reduces infrastructure footprint and operational expenses
- Self-healing networking: InfiniBand fabric recovers from link failures 5,000× faster than software-based solutions
- Flexible deployment: Backward compatible with HDR and EDR; supports modern topologies like Dragonfly+, SlimFly, and 6D torus
- Enhanced telemetry: With UFM, gain real-time network insights and AI-driven analytics for proactive maintenance
- Standard cooling: P2C airflow matches typical data center configurations with front cold-aisle setups
Buyer Checklist
- Confirm airflow direction (P2C front-to-rear for this model)
- Verify power input voltage (100-127V / 200-240VAC)
- Choose router (MQM9790) or switch (MQM9700) based on routing needs
- Select appropriate QSFP56 cables or optics
- Check rack depth compatibility (590.6mm / 27")
- Plan for 1U spacing and cooling clearance
- Design routing topology for multi-subnet deployments
Precautions
- Ensure input voltage range (100-127VAC or 200-240VAC) matches your facility power before connecting
- Allow sufficient clearance for front-to-rear airflow; do not block fan exhaust at the rear or intake at the front
- Operating temperature must be maintained between 0°C and 45°C; relative humidity 10%-85% non-condensing
- Use only approved QSFP56 cables and optical modules to ensure signal integrity and compliance
- Verify that your rack cooling configuration supports P2C (front-to-rear) airflow before installation
- For routing configurations, consult the InfiniBand Fabric Design Guide for proper topology planning
Frequently Asked Questions
What is the difference between MQM9790-NS2F and MQM9700-NS2F?
The MQM9790 is an InfiniBand router with advanced routing capabilities, while the MQM9700 is a standard switch. Both share the same hardware platform and airflow direction, but the MQM9790 supports subnet routing, traffic isolation, and multi-tenant segmentation.
What is the difference between MQM9790-NS2F and MQM9790-NS2R?
The only difference is the airflow direction. NS2F has P2C (front-to-rear) airflow, while NS2R has C2P (rear-to-front) airflow. All other specifications, including routing capabilities, are identical.
What routing capabilities does the MQM9790 provide?
The MQM9790 series supports InfiniBand subnet routing, enabling communication between different subnets, traffic isolation for multi-tenant environments, and advanced forwarding for large-scale fabrics with thousands of nodes.
What cables are supported?
QSFP56 connectors support passive copper DAC, active optical cables (AOC), and optical transceivers. For specific SKUs, refer to the NVIDIA LinkX cable family.
Can this router be used with HDR or EDR adapters?
Yes, the Quantum-2 series is backward compatible with HDR (200Gb/s) and EDR (100Gb/s) InfiniBand technologies.
What is the maximum number of nodes supported?
With advanced routing capabilities, the MQM9790 can support fabrics with tens of thousands of nodes across multiple subnets. The integrated subnet manager can manage up to 2,000 nodes per subnet.
How do I choose between NS2F and NS2R?
Choose NS2F (front-to-rear) if your rack has cold air intake at the front; choose NS2R (rear-to-front) if your rack has cold air intake at the rear. Always match the airflow direction of adjacent equipment.
Service & Support
Hong Kong Starsurge Group Co., Limited provides global delivery and multilingual support for the MQM9790-NS2F. Our services include:
- Pre-sales consultation on routing topology design and compatibility
- Firmware updates and technical documentation
- Warranty support and RMA handling
- Custom integration and system deployment assistance
About Hong Kong Starsurge Group Co., Limited
Hong Kong Starsurge Group Co., Limited is a technology-driven provider of network hardware, IT services, and system integration solutions. Founded in 2008, the company serves customers worldwide with products including network switches, NICs, wireless access points, controllers, cables, and related networking equipment.
Backed by an experienced sales and technical team, Starsurge supports industries such as government, healthcare, manufacturing, education, finance, and enterprise. The company also offers IoT solutions, network management systems, custom software development, multilingual support, and global delivery. With a customer-first approach, Starsurge focuses on reliable quality, responsive service, and tailored solutions that help clients build efficient, scalable, and dependable network infrastructure.
Want to Know more details about this product