{"resourceId":"nps-dgx-gb300-power-cooling-specs-2026","versions":[{"version":"external-7c313db30ef4d7c733c30cd6f948ff8afb8c4b82bb4e4172a242b3fd9a0777fb","resource":{"id":"nps-dgx-gb300-power-cooling-specs-2026","title":"NPS documents GB300 peak power (~135 kW) and closed-loop liquid cooling inside existing capacity","organization":"Naval Postgraduate School AI Task Force","sector":"Defense higher education","geography":"United States (Monterey, California)","publishedAt":"2026 (page current at retrieval; system delivered 2026)","publicationDate":null,"eventDate":null,"sourceName":"NPS AI Task Force GB300 page","sourceLabel":"Government / institutional technical summary","sourceUrl":"https://nps.edu/web/ai/gb300","evidenceClass":"standards-guidance","outcomeClass":"cautionary","topics":["infrastructure","governance-procurement","operating-model"],"finding":"NPS publishes operator-facing facts for the donated DGX GB300: 72 Blackwell Ultra GPUs, 36 Grace CPUs, ~135 kW peak compute draw with expected daily operations at 50–80% of peak, and dual closed-loop liquid cooling that rejects heat via roof cooling water. The page states the platform operates within existing power and water capacities.","sledRelevance":"Gives SLED facility and procurement teams quantitative planning inputs for Blackwell rack-scale education/research deployments, and clarifies that within-existing-capacity claims still require local site verification.","evidence":"NPS GB300 page lists architecture (18 trays x 4 GPU/2 CPU), FP4 performance, 20 TB HBM, use-case list (weather, cyber, LLM fine-tuning, multi-agent tactics), and Power and Cooling section with 135 kW peak and 50–80% anticipated daily operations.","architectureImplications":"Budget liquid cooling CDUs/heat exchangers, dual-loop facility water, and rack support gear as first-class design objects alongside GPUs. Size power distribution for ~135 kW compute peaks plus support loads.","governanceImplications":"Shared research capability with approved collaborations requires written access criteria spanning faculty, students, labs, and industry partners aligned to defense priorities.","securityPrivacyImplications":"Use cases include synthetic data for denied/classified sparsity and adversarial resilience training—implying strong data classification and experiment isolation practices.","caveats":"Institutional specifications and anticipated utilization bands are not independent measurements. Performance figures (PFLOPS) are product specs, not NPS workload benchmarks.","streamIds":["nvidia","research","campus-operations"],"roles":{"sales":"Use the published 135 kW peak and liquid-cooling design to ground facility conversations early; avoid selling GPUs without a power/cooling readiness story.","engineering":"Validate facility water/power against peak and duty-cycle assumptions; design closed-loop liquid cooling and support-rack thermal paths explicitly.","delivery":"Commissioning must include fluid management, duty-cycle monitoring, and shared-access runbooks tied to the stated 50–80% expected operating band."},"roleDetails":{"sales":{"customerProblem":"Campus/agency rooms often lack documented headroom for ~100+ kW liquid-cooled AI racks.","stakeholders":["facilities","CIO","HPC lead","EHS","procurement"],"discoveryQuestions":["What is measured power/water headroom at the proposed room?","Who owns liquid-cooling maintenance?","What duty cycle is planned for the first 90 days?"],"valueHypothesis":"Published NPS figures provide a concrete planning envelope for Blackwell education deployments when verified locally.","potentialEngagement":"Joint facilities+HPC assessment against the NPS-published envelope before proposal commitment.","unsupportedClaims":["Any site can host GB300 without upgrades","135 kW covers full support infrastructure"]},"engineering":{"fit":"Fit when site can sustain ~135 kW compute peaks with closed-loop liquid cooling and documented water rejection path.","architectureIntegration":"Map compute trays, NVLink switches, CDUs/heat exchangers, and facility chilled water into one design package.","prerequisites":["Accurate one-line diagrams","Coolant chemistry/monitoring","Seismic/anchoring as required","Duty-cycle telemetry"],"constraints":["Peak vs average draw differs","Support components add thermal load","Existing rooms may need upgrades despite within-capacity claims elsewhere"],"security":"Isolate research datasets and experiment tenancy on shared system.","proofOfValue":["Measured peak draw within design","Cooling loop stability under load","Documented tray/switch inventory"]},"delivery":{"work":"Facility conversion, loop commissioning, and operator training for liquid-cooled operations.","dependencies":["mechanical/electrical contractors","OEM partners","HPC ops","EHS"],"ownership":"Facilities owns plant; HPC owns compute operations; joint incident response.","skillsAdoption":"Operators need liquid-cooling and high-density power runbooks, not only CUDA skills.","governanceCheckpoints":["Pre-energization checklist","Leak/response drill","30-day duty-cycle review"],"acceptanceCriteria":["System runs within published power band","Cooling alarms exercised","Shared access SOP live"],"risks":["Underestimated support power","Cooling water constraints","Operator skill gaps"]}},"retrievedAt":"2026-09-15T14:15:00Z","enrichedAt":"2026-09-15T14:17:00Z","enrichmentBasis":"retrieved source","accessibilityWorkforceImplications":"Shared capability expands who can train on frontier systems, but only if intake and training paths are funded.","procurementImplications":"Require vendors to price power, cooling, and support racks with the compute BOM; treat 135 kW as a minimum planning input to verify.","operatingModelImplications":"Expected 50–80% duty cycle implies capacity planning and chargeback/fair-share policies for campus and partner users.","sourceVerification":{"openedUrl":"https://nps.edu/web/ai/gb300","referenceExcerpt":"At peak use, the DGX GB300’s compute components draw roughly 135 (kilowatts) kW.","promptVersion":"sled-research-v4.0","model":null,"basis":"agent-reported inspection"}}}]}