The DRACO GB200 NVL36 Rack-Scale AI System is a turnkey, liquid-cooled single-rack (nvlink domain) AI training cluster delivering 6,912 GB HBM3e of aggregate HBM3e in a single unified NVLink domain. It arrives racked, cabled, cooled and validated \u2014 ready to train and serve the largest models on-premises, in INR, on a GST invoice.<\/p>\n
Engineered for organisations building serious in-house AI capacity, it removes the integration risk of assembling a cluster yourself: RDP sizes the NVLink domain, fabric, storage, power and cooling as one validated system delivered as a single SKU with one warranty and one support contract.<\/p>\n
This sits at the cluster-scale train-and-serve<\/strong> stage: a complete, validated AI system rather than a single server. With 6,912 GB HBM3e of HBM3e in one NVLink domain, it is sized to sustain<\/strong> real distributed training and large-scale serving \u2014 the owned alternative to a cloud cluster where the meter never stops.<\/p>\n We don’t publish inflated peak numbers. The honest picture: 6,912 GB HBM3e of HBM3e across 36 Blackwell GPUs in one NVLink domain is sized for Trillion-scale-class distributed training and large-scale serving. Want certainty? Request a free benchmark of your model and dataset \u2014 including a scaling test \u2014 on this exact configuration before you buy<\/strong>; we’ll send back real tokens\/sec, scaling efficiency and timings.<\/p>\n For a sustained training cluster, owning beats renting decisively: an always-on cloud cluster of this size is the dominant line in an AI budget, and on-prem removes egress fees and keeps data and weights in-house. RDP pricing is fixed in INR with a GST input-credit-eligible invoice \u2014 ask for a 3-year cluster TCO comparison<\/strong>.<\/p>\n Ships pre-integrated and validated<\/strong>: NVIDIA driver, CUDA, cuDNN, NCCL, the InfiniBand stack, Docker and NVIDIA Container Toolkit, with Slurm or Kubernetes, PyTorch and vLLM \/ Triton \/ TensorRT-LLM on Ubuntu LTS. Optional managed cluster operations, scheduler and observability setup.<\/p>\n A single-rack (nvlink domain) liquid-cooled system \u2014 plan facility power, CDU\/manifold and water, and the InfiniBand spine. RDP scopes power, cooling and floor\/rack requirements as part of the design. (Exact PSU\/PDU ratings, BTU, flow and facility figures confirmed on the build sheet.)<\/em> Full out-of-band management across the system.<\/p>\n 14 years of Make-in-India infrastructure and 300,000+ devices shipped<\/strong>. Indian OEM, INR pricing, GST tax invoice (HSN 8471), pan-India onsite engineers, GeM availability, and DPDP \/ sovereign-AI-ready deployment.<\/p>\n This is a turnkey rack-scale AI system, made to order \u2014 talk to an RDP solution architect<\/strong>, size the NVLink domain, fabric and facility, get a 3-year TCO, and benchmark your own model with a scaling test before you commit.<\/strong> Request a quote to begin.<\/p>\n","protected":false},"excerpt":{"rendered":" GB200 NVL36 \u00b7 18\u00d7 NVIDIA Grace (ARM) \u00b7 Grace LPDDR5X coherent memory \u00b7 240 TB NVMe \u00b7 Single-Rack (NVLink domain) \u00b7 liquid-cooled<\/p>\n","protected":false},"featured_media":2093,"comment_status":"open","ping_status":"closed","template":"","meta":{"_yoast_wpseo_title":"","_yoast_wpseo_metadesc":"","rank_math_title":"DRACO GB200 NVL36 Rack-Scale AI System \u2014 6,912 GB HBM3e HBM, unified NVLink | RDP GPU Mart","rank_math_description":"Turnkey single-rack (nvlink domain) AI cluster \u2014 GB200 NVL36 (6,912 GB HBM3e), unified NVLink domain, liquid-cooled. Train Trillion-scale-class models on-prem. Make-in-India, GST invoice. Request a quote.","_hermes_jsonld":""},"product_brand":[],"product_cat":[17],"product_tag":[83],"class_list":["post-68","product","type-product","status-publish","has-post-thumbnail","product_cat-rack-scale-ai-systems","product_tag-ready-to-buy","pa_form-factor-rack","pa_gpu-model-nvidia-gb200-grace-blackwell","pa_industry-media-gaming-entertainment","pa_industry-neocloud","pa_industry-public-sector-sovereign-ai","pa_industry-research-higher-education","pa_series-draco","pa_use-case-agentic-ai","pa_use-case-computer-vision","pa_use-case-generative-ai","pa_use-case-hpc-ai","pa_use-case-inference","pa_use-case-llm-training","pa_use-case-fine-tuning","pa_use-case-nlp-speech","pa_use-case-rag","pa_use-case-sovereign-ai","pa_workload-fit-unified-nvlink-domain-training-rack","first","onbackorder","taxable","shipping-taxable","product-type-simple"],"_links":{"self":[{"href":"https:\/\/rdp.in\/gpu-mart\/wp-json\/wp\/v2\/product\/68","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/rdp.in\/gpu-mart\/wp-json\/wp\/v2\/product"}],"about":[{"href":"https:\/\/rdp.in\/gpu-mart\/wp-json\/wp\/v2\/types\/product"}],"replies":[{"embeddable":true,"href":"https:\/\/rdp.in\/gpu-mart\/wp-json\/wp\/v2\/comments?post=68"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/rdp.in\/gpu-mart\/wp-json\/wp\/v2\/media\/2093"}],"wp:attachment":[{"href":"https:\/\/rdp.in\/gpu-mart\/wp-json\/wp\/v2\/media?parent=68"}],"wp:term":[{"taxonomy":"product_brand","embeddable":true,"href":"https:\/\/rdp.in\/gpu-mart\/wp-json\/wp\/v2\/product_brand?post=68"},{"taxonomy":"product_cat","embeddable":true,"href":"https:\/\/rdp.in\/gpu-mart\/wp-json\/wp\/v2\/product_cat?post=68"},{"taxonomy":"product_tag","embeddable":true,"href":"https:\/\/rdp.in\/gpu-mart\/wp-json\/wp\/v2\/product_tag?post=68"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}Industry use cases<\/h3>\n
\n
Performance \u2014 and how to be sure<\/h3>\n
Series & upgrade path<\/h3>\n
\n
On-prem vs cloud \u2014 the TCO case<\/h3>\n
Software & day-one readiness<\/h3>\n
Power, cooling & rack integration<\/h3>\n
Deployment, warranty & support<\/h3>\n
\n
Why RDP<\/h3>\n
Buy with confidence<\/h3>\n