Skip to main content

Use case

Enterprise Private AI Factory

Private enterprise AI / RAG / inference.

A controlled environment for enterprise models, retrieval, and inference, inside the organization's security and data boundary.

Architecture

  1. 01Data Sources
  2. 02Ingestion
  3. 03Object / File Storage
  4. 04Vector / RAG Layer
  5. 05Compute
  6. 06Inference Platform
  7. 07Enterprise Applications
  8. 08Operate

This path is one instance of the factory data pipeline: data to workload, landed on the accelerator plant, then operate. NeuronPlant engineers the cell and the vendor coordination. We do not write every model. We do not construct the hall.

Applications / Platform →Runtime / Operate →

NP-STO / What to look for

Storage is a workload decision.

Throughput numbers below for training come from the NVIDIA SuperPOD with NVIDIA DGX B300 reference architectures. RAG and NIM notes follow how those stacks actually move bytes, plus the NVIDIA NIM support matrix for profile sizing.

Training / post-training

Sustained sequential GB/s and checkpoint burst, not IOPS marketing.

Shared parallel filesystem for datasets. Separate checkpoint target that can absorb a simultaneous flush from the job. NVIDIA SuperPOD B300 RA states high-performance storage I/O per node must exceed 40 GB/s (Ethernet RA) and, on the Quantum-X800 RA, 80 GB/s.

Typical: HPS on IB (QM9700) or RoCE (SN5600) as the storage/data fabric, plus a quieter NFS/user tier on SN5610 as North-South / front-end. Do not fold HPS into a generic East-West compute line if dedicated.

Avoid: A single enterprise NAS asked to be both home directories and the training scratch.

RAG / enterprise retrieval

Random read IOPS, metadata, and a clean split between documents and the vector index.

Object or file store for source documents. NVMe-backed or in-memory vector index. Ingest path sized to nightly rebuilds, not to training checkpoints.

Typical: Object + vector engine. North-South / front-end Ethernet is usually enough if the index is local or on a low-latency tier. That is not the East-West compute fabric.

Avoid: Putting the vector store on the same parallel FS as a training job.

Inference / NIM serving

Fast model-weight load, local NVMe or a warm cache, isolation from training I/O, and a KV / prefix cache plan on large serving.

NVIDIA NIM selects a profile from its support matrix (precision, tensor-parallel size, LoRA). Weights load from local NVMe. That is not the whole inference memory story. On a large workload, KV cache needs a hierarchy: GPU HBM, host overflow, and sometimes a shared prefix cache across the serving plane. That tier is not a storage array and not a checkpoint pool.

Typical: Node-local NVMe for NIM images and weights. KV in HBM first. Host memory and a serving-plane prefix cache when context and concurrency fill the GPU.

Avoid: Loading 70B-class weights over a contended NFS home share. Treating KV as another LUN on the training filesystem.

Research / HPC mix

Scratch that can be purged, project space that cannot, and a scheduler that knows the difference.

Burst scratch on the HPS fabric. Project datasets on a capacity tier. Home and logs on the user-storage Ethernet path, as in the NVIDIA SuperPOD split.

Typical: Two storage systems: HPS + user storage. NVIDIA documents this split explicitly.

Avoid: One quota, one protocol, every lab on the same queue.

NP-CBL / Fabric planes

East-West · North-South · Management. Cabling builds the plants.

DWG-CBL-01 / NOT FOUR EQUAL JOBSSTRUCTURED CABLING BUILDS THESE PLANTSENTERPRISE / USERS / APPLICATIONSAPI GATEWAY / LB · FIREWALLS · CORENORTH-SOUTH / FRONT-END FABRICFRONT-END / APP / ENTERPRISEAI PLATFORM / DGX ENVIRONMENTNOT THE HALL · THE ACCELERATOR CELLEAST-WEST COMPUTE · QUANTUM-X / SPECTRUM-XGPU-TO-GPUDGX-TO-DGX · RDMASTORAGE / DATA · DEDICATED · COMPUTE ↔ HPSMGMT / OOB · NOT N-SDGXGPU rail 1DGXGPU rail 2DGXGPU rail 3DGXGPU rail 4OVERHEAD BASKETS / HOW THE PLANTS ARE BUILT / NOT A FIFTH FABRICEW · TRAY A · COMPUTE MPO · RAIL-ALIGNEDEW · TRAY B · STORAGE / DATA · OFF THE COMPUTE RAILSNS · TRAY C · FRONT-ENDMGMT · TRAY D · OOB / BMCN-S IS USERS AND APPSE-W IS GPU / DGX SCALE-OUTUNDERFLOOR = POWER / LIQUID · OVERHEAD = FIBER · SLACK AT RACK AND FRAMESTORAGE AND OOB INDEPENDENT

DWG-CBL-01 · laying out

EW / East-West

Hall-internal scale-out. Compute fabric first. Storage/data fabric on its own plane when dedicated. Not user traffic.

01 / East-West Compute Fabric

MPO/MTP trunks, 400/800G optics, rail-aligned

GPU-to-GPU, DGX-to-DGX, scale-out. RDMA/RoCE or InfiniBand. NVIDIA Spectrum-X or Quantum-X as the high-speed EW compute fabric. NCCL, all-reduce. One polarity map. One tray.

02 / Storage / Data Fabric

Separate MPO or LC plant onto the storage leaf when dedicated

Compute to high-performance storage: datasets and checkpoints. Own internal data plane when physically separate. Do not fold it into a generic East-West compute line. Do not auto-label it North-South or East-West without that context. NVIDIA SuperPOD keeps this off the compute rails.

NS / North-South

Users, applications, API gateway/load balancer, enterprise/core, firewalls. Application-facing Ethernet. Not another GPU rail.

03 / North-South / Front-End Fabric

LC or lower-count MPO, 100G-class typical

Users, apps, API gateway/LB, enterprise/core, application-facing Ethernet. Not just another rail next to the GPU rails. Must not share a tray or a cassette with training I/O.

MGMT / Management

BMC, management, operational control. Never classify as application North-South.

04 / Management / OOB Fabric

Cat6A or dedicated 1/10G optics. Own patch field.

BMC, management, operational control. Lights-out, firmware, serial, leak sensors. Never application North-South. If this plant dies inside a compute bundle, you cannot even see why the hall is dark.

Infrastructure

Site
Inside the organization's security and data boundary.
Power
Rack-level redundancy, typically 20-40 kW/rack
Cooling
Air or targeted liquid depending on density
Compute
Accelerators for NIM / RAG serving. GPU plant: NVIDIA DGX B200 or NVIDIA DGX B300, or HGX. TPU and other ASICs when that is the job.
Network
Spectrum-X East-West compute fabric, dedicated storage/data path, separate North-South front-end
Structured cabling
Labeled LC/MPO plant. Management/OOB on its own patch field. Factory-tested trunks. Polarity map before first optic.
Storage
Object for documents + NVMe or memory vector index. Model weights on local NVMe.
Platform
Sources → ingest → RAG / inference → enterprise applications.
Operate
Docker / Kubernetes cell. Observe as part of the plane.

Design considerations

  • Dataset size and retention
  • Growth of concurrent sessions
  • Availability target
  • Security and data residency
  • Latency to enterprise applications
  • Data lifecycle and audit
  • Whether the existing hall plant can take 400G without a recable
  • How the serving plane is observed after handover

NP / Intake

Tell us what you need to achieve with AI.

One accountable delivery partner from AI requirement to a delivered factory: site, power, cooling, compute, fabric, structured cabling & passive infrastructure, storage, platform, procurement, and handover. You do not need a vendor list or a hall name first. The hall is a supplier.

Start a Project →