QUASAR Spectrum-X 800G Ethernet AI Fabric Kit

Overview

The QUASAR Spectrum-X 800G Ethernet AI Fabric Kit is a turnkey Ethernet AI fabric kit — a high-bandwidth Spectrum-X switch plus BlueField-3 DPUs, providing 32× 800G + BlueField-3 DPUs at 25.6 Tb/s over Spectrum-X Ethernet (RoCE) + BlueField-3 DPU, with Low-latency RoCE. It is the interconnect that turns GPU servers into a cluster, carrying the collective (all-reduce) traffic that dominates multi-node AI training — on-premises, in INR, on a GST invoice.

Engineered as the fabric layer of an RDP AI cluster, it is delivered configured, cabled and validated alongside the compute and storage it connects, with DPUs that offload networking, storage and security from the host CPUs.

Key highlights

• 32× 800G + BlueField-3 DPUs · 25.6 Tb/s — high-bandwidth switching plus DPUs to wire a GPU cluster at full bandwidth.
• Spectrum-X Ethernet (RoCE) + BlueField-3 DPU — RoCE with adaptive routing and congestion control, plus DPU offload for networking/storage/security.
• Low-latency RoCE — low latency so distributed training scales with near-linear efficiency.
• 800G / 400G per port — matched to your GPU NICs/DPUs.
• DPU-accelerated — BlueField-3 offloads host CPUs and isolates tenants.
• Validated with RDP compute & storage — part of a tested cluster.
• On-prem sovereignty — your fabric, your data path; DPDP-friendly, air-gappable.
• Make-in-India OEM — predictable INR pricing, GST tax invoice (HSN 8517), pan-India onsite support, GeM-procurable.

Where it fits

• GPU cluster fabric (primary): the interconnect that carries distributed-training collectives between GPU servers.
• Storage fabric: high-bandwidth path between GPU nodes and the NVMe parallel-FS (GPUDirect).
• Multi-tenant isolation: DPUs isolate and secure tenants on a shared fabric.
• Ethernet-standardised AI: for teams standardising on Ethernet/RoCE.

How it works

The Spectrum-X switch runs RoCE over Ethernet with adaptive routing and congestion control, while BlueField-3 DPUs in each host offload networking, storage and security — keeping the fabric lossless and the host CPUs free — so all-reduce traffic, which dominates multi-node training, completes fast. RDP sizes the topology (oversubscription, spine count, DPU count) to your cluster. Honest note: real scaling efficiency depends on topology, collective library and model — we validate it on your cluster, not just a peak number.

Industry use cases

• AI/ML & foundation-model teams — the fabric that lets training scale across nodes.
• Government & national labs — sovereign HPC/AI interconnect.
• Neocloud / AI providers — multi-tenant fabric for a GPU cloud.
• BFSI & research — low-latency clusters for training and HPC.
• Telecom — high-bandwidth AI fabric at core/edge.
• Manufacturing & energy — HPC + AI cluster interconnect.

Performance — and how to be sure

We don’t publish inflated peak numbers. The honest picture: 25.6 Tb/s at Low-latency RoCE is sized to carry cluster collectives without becoming the bottleneck. Want certainty? Request a free scaling test of your training job across nodes on this fabric before you buy; we’ll send back real all-reduce bandwidth and scaling efficiency.

Series & scale path

• QUASAR (performance fabric tier) — this.
• Speed ladder: 200G Ethernet → 400G NDR → 800G XDR InfiniBand.
• Scale: single kit → multi-switch fabric for thousands of GPUs.

On-prem vs cloud — the TCO case

An owned fabric is a one-time asset that removes the per-node networking premium of cloud GPU clusters and keeps the data path in-house. RDP pricing is fixed in INR with a GST input-credit-eligible invoice — ask for a fabric design and TCO for your cluster.

Software & integration

Integrates with the AI stack: NCCL over RoCE, NVIDIA Spectrum-X and DOCA for the DPUs, with fabric monitoring and telemetry. Validated with RDP GPU servers, rack-scale systems and storage.

Power, cooling & rack integration

A Switch + DPU kit with redundant PSUs and front-to-back airflow — specify rack power, cooling and cable plant (DAC/AOC/transceivers). (Exact power draw, BTU and cabling confirmed on the build sheet.) Full out-of-band management.

Deployment, warranty & support

• Made to order, configured, cabled and validated in India as part of the cluster; lead time confirmed at quote.
• In the box: switch, BlueField-3 DPUs, rails, cables; transceivers/cables scoped at quote.
• Onsite warranty + AMC with pan-India coverage and an RMA/escalation path (exact terms confirmed at quote).

Why RDP

14 years of Make-in-India infrastructure and 300,000+ devices shipped. Indian OEM, INR pricing, GST tax invoice, pan-India onsite engineers, GeM availability, and DPDP / sovereign-AI-ready deployment.

Buy with confidence

This is an AI cluster fabric, made to order — talk to an RDP solution architect, design the topology and oversubscription for your cluster, get a fabric TCO, and run a scaling test before you commit. Request a quote to begin.

Specifications