NVIDIA Quantum-X800 InfiniBand Switch · Q3200 Series

NVIDIA Q3200-RA 800Gb/s InfiniBand Switch, 2U Dual 36-Port, 36 OSFP, Air-Cooled

The Q3200-RA is a 2U air-cooled system from the NVIDIA Quantum-X800 series that houses two independent 800Gb/s InfiniBand switches in a single enclosure. Each switch provides 36 non-blocking 800Gb/s ports across 18 OSFP connectors and 28.8 Tb/s throughput, for 72 ports and 57.6 Tb/s combined. It is well-suited for connecting new compute clusters to previous-generation Quantum and Quantum-2 InfiniBand storage infrastructure.

B2B supply and project configuration support provided by Hong Kong Starsurge Group Co., Limited. Availability, lead time and pricing are confirmed per order.

Product Overview

The Q3200-RA is part of the NVIDIA Quantum-X800 family of 800Gb/s InfiniBand switches, which deliver 800Gb/s throughput with ultra-low latency and advanced NVIDIA In-Network Computing, positioned for trillion-parameter-scale generative AI. Unlike the single-switch 4U Q3400-RA, the Q3200-RA is a 2U air-cooled system that houses two independent switches within a single enclosure, each providing 36 ports at 800Gb/s across 18 OSFP connectors.

The platform incorporates remote direct-memory access (RDMA), fourth-generation NVIDIA Scalable Hierarchical Aggregation and Reduction Protocol (SHARP), adaptive routing, telemetry-based congestion control, self-healing and Quantum-X Photonics co-packaged optics (CPO) technologies. These features elevate overall application performance within HPC and AI data centers.

The Q3200 fixed-configuration switches are well-suited for connecting new compute clusters to previous-generation Quantum and Quantum-2 InfiniBand storage infrastructure. All Quantum-X800 switches include a dedicated OSFP InfiniBand in-band management port for NVIDIA UFM, separated on the front panel from the data ports, allowing the full set of standard ports to be used for data network connectivity and simplifying port allocation and topology design.

Key Features

• Two independent 800Gb/s switches in one 2U enclosure, each with 36 non-blocking 800Gb/s ports and 28.8 Tb/s throughput.
• Two groups of 18 OSFP connectors, for 36 OSFP connectors per system and 72 × 800Gb/s ports in total.
• Ultra-low latency and 200Gb/s per-lane SerDes technology for high-performance AI and HPC fabrics.
• Fourth-generation NVIDIA SHARP with FP8 precision and new collective operations such as ReduceScatter and ScatterGather.
• Enhanced adaptive routing and telemetry-based congestion control for nearly perfect effective bandwidth and performance isolation.
• Quantum-X Photonics CPO technology at the platform level, reducing total power consumption, latency and TCO while enhancing resiliency and serviceability.
• NVIDIA NVOS software for chassis management, with CLI, REST APIs, SNMP and gNMI telemetry.
• Dedicated OSFP 400Gb/s InfiniBand in-band management port for UFM, separated from data ports.
• Compact 2U air-cooled form factor for standard 19-inch racks, with 200–240 VAC power feed and power capping.

Technology

800Gb/s InfiniBand for Trillion-Parameter AI

NVIDIA Quantum-X800 InfiniBand switches deliver 800Gb/s throughput with ultra-low latency and advanced NVIDIA In-Network Computing, which is essential for handling trillion-parameter-scale generative AI. Quantum-X800 switches enable 2x faster speeds and 5x higher scalability for AI compute fabrics, with two-tier fat-tree topologies that can support over 10,000 800Gb/s host connections.

Fourth-Generation NVIDIA SHARP

Quantum-X800 switches support SHARP for efficient offloading of compute operations to the network, boosting performance by up to 9x. The fourth generation of SHARP adds support for FP8 precision and new collective operations, such as ReduceScatter and ScatterGather, extending in-network computing to a broader set of AI workloads.

Adaptive Routing and Telemetry-Based Congestion Control

Quantum-X800 switches support enhanced adaptive routing and telemetry-based congestion control. This enables nearly perfect effective bandwidth, as well as performance isolation for multi-tenant and multi-job environments, helping operators maintain predictable application performance across shared fabrics.

Dual Independent Switches and Legacy Interconnect

The Q3200-RA houses two independent switches within a single 2U enclosure, each providing 36 ports at 800Gb/s. The Q3200 fixed-configuration switches are well-suited for connecting new compute clusters to previous-generation Quantum and Quantum-2 InfiniBand storage infrastructure, helping operators bridge generations without redesigning the entire fabric.

Management with NVOS and UFM

Quantum-X800 switches include NVIDIA Networking OS Software (NVOS) for comprehensive chassis management and system configuration. NVOS supports a command-line interface (CLI), REST APIs, Simple Network Management Protocol (SNMP) and gRPC Network Management Interface (gNMI) telemetry. A dedicated OSFP InfiniBand in-band management port for NVIDIA UFM is separated on the front panel from the data ports, allowing the full set of standard ports to be used for data network connectivity and streamlining topology design.

Typical Deployments

• New compute clusters connecting to previous-generation Quantum and Quantum-2 storage infrastructure: the Q3200 fixed-configuration switches are well-suited for this mixed-generation role.
• Smaller-scale AI and HPC platforms: two independent 36-port switches in one 2U enclosure for compact fabric building blocks.
• Trillion-parameter generative AI clusters: fabrics that require 800Gb/s per port and in-network computing.
• Multi-tenant and multi-job environments: performance isolation and nearly perfect effective bandwidth.
• Standard 19-inch air-cooled racks: the Q3200-RA is an air-cooled system with 200–240 VAC power feed.
• Government, healthcare, manufacturing, education, finance and enterprise: organizations building or extending AI and HPC infrastructure.

Compatibility

The Q3200-RA is an NVIDIA Quantum-X800-series 800Gb/s InfiniBand system with two independent 36-port switches in a 2U enclosure, for a total of 72 × 800Gb/s ports across two groups of 18 OSFP connectors. It is an air-cooled system managed through NVIDIA NVOS and UFM. Compatibility depends on the specific fabric design, cable plant, optics and host channel adapters in your environment.

• Two groups of 18 OSFP connectors; pluggable optics connectivity. Exact optics and cable compatibility should be confirmed against the NVIDIA compatibility list for your deployment.
• Dedicated OSFP 400Gb/s InfiniBand in-band management port for UFM, separated from data ports.
• NVOS software with CLI, REST APIs, SNMP and gNMI telemetry.
• Power feed: 200–240 VAC; confirm site power and rack clearance before ordering.
• Designed for connecting new compute clusters to previous-generation Quantum and Quantum-2 InfiniBand storage infrastructure; confirm mixed-generation fabric design before ordering.

For fabric-level compatibility, optics qualification and firmware matching, please confirm with our technical team before ordering.

Specification Table

Source: NVIDIA Quantum-X800 Q3200-RA datasheet provided by the customer. Parameters not publicly specified in the source are marked accordingly.

Selection Guide

The Quantum-X800 family includes air-cooled and liquid-cooled systems with different form factors, connectivity and power options. Use the table below to match rack type, cooling, port density and connectivity to your data center design.

Selection note: the Q3200-RA is the compact 2U air-cooled option with two independent 36-port switches, suited to smaller-scale platforms and mixed-generation connectivity. For a single 144-port 800Gb/s switch in 4U, consider Q3400-RA; for DC power, Q3401-RD; for liquid-cooled CPO, Q3450-LD.

Advantages

• Compact dual-switch design: two independent 36-port 800Gb/s switches in a single 2U enclosure, for 72 ports and 57.6 Tb/s combined throughput.
• Mixed-generation connectivity: well-suited for connecting new compute clusters to previous-generation Quantum and Quantum-2 InfiniBand storage infrastructure.
• High AI performance: 800Gb/s per port, ultra-low latency and fourth-generation SHARP for in-network computing with FP8 precision.
• Predictable multi-tenant performance: enhanced adaptive routing and telemetry-based congestion control deliver nearly perfect effective bandwidth and performance isolation.
• Simplified topology design: a dedicated OSFP in-band management port for UFM is separated from data ports, so the full set of standard ports is available for data connectivity.
• Operational efficiency: 2U air-cooled form factor for standard 19-inch racks, NVOS management with CLI, REST APIs, SNMP and gNMI telemetry, and power capping.

Service & Support

Hong Kong Starsurge Group Co., Limited supports this product with an experienced sales and technical team. Support scope can include pre-sales configuration review, rack and power confirmation, compatibility checks, and order follow-up.

• Pre-sales selection support for form factor, cooling, port density, connectivity and power feed.
• Compatibility confirmation for OSFP optics, cables and host adapters before ordering.
• Mixed-generation fabric review for connecting new clusters to Quantum / Quantum-2 storage infrastructure.
• Multilingual support and global delivery available through Starsurge.

FAQ

What is the Q3200-RA?

It is a 2U air-cooled system in the NVIDIA Quantum-X800 series that houses two independent 800Gb/s InfiniBand switches, each with 36 non-blocking 800Gb/s ports across 18 OSFP connectors and 28.8 Tb/s throughput.

How many ports and what throughput does the system provide?

Two independent switches, each with 36 × 800Gb/s non-blocking ports and 28.8 Tb/s throughput. Combined, the system provides 72 × 800Gb/s ports and 57.6 Tb/s throughput across two groups of 18 OSFP connectors.

What is the difference between Q3200-RA and Q3400-RA?

The Q3200-RA is a 2U air-cooled enclosure housing two independent 36-port switches (72 ports total), while the Q3400-RA is a 4U air-cooled single switch with 144 ports and 72 OSFP connectors. The Q3200-RA is well-suited to smaller-scale platforms and mixed-generation connectivity; the Q3400-RA is a higher-density single-switch option.

Is the Q3200-RA suitable for connecting to previous-generation InfiniBand?

The Q3200 fixed-configuration switches are described as well-suited for connecting new compute clusters to previous-generation Quantum and Quantum-2 InfiniBand storage infrastructure. Confirm the specific mixed-generation fabric design and firmware compatibility before ordering.

Does it support fourth-generation SHARP?

Yes. Quantum-X800 switches support SHARP for efficient offloading of compute operations to the network, boosting performance by up to 9x. The fourth generation adds FP8 precision and new collective operations such as ReduceScatter and ScatterGather.

What management software does it use?

It uses NVIDIA Networking OS Software (NVOS) for chassis management and configuration, with CLI, REST APIs, SNMP and gNMI telemetry. A dedicated OSFP 400Gb/s InfiniBand in-band management port supports NVIDIA UFM.

Can you confirm compatibility with my optics and cables?

Yes. Please share your planned OSFP optics, cables and host adapter models, and our technical team will confirm compatibility before ordering.

Precautions