{"resourceId":"cisco-enterprise-ai-factory-gpu-utilization-onprem-security-20261106","versions":[{"version":"external-58dfb07b67e39f49be1f9cb31110fd9ab712b3e6b7a5c98958904c5cee1e4fb5","resource":{"id":"cisco-enterprise-ai-factory-gpu-utilization-onprem-security-20261106","title":"Cisco IT says its multi-tenant enterprise AI factory tripled GPU utilization to ~50% and hosts MyAgent plus Cisco IQ at 99.99%+ SLA","organization":"Cisco","sector":"Vendor IT (Cisco-on-Cisco) AI infrastructure, security and observability; transferable to SLED estates with caveats","geography":"Global (Cisco enterprise data centers); U.S. SLED transferability limited","publishedAt":"November 6, 2026","publicationDate":"2026-11-06","eventDate":null,"sourceName":"Cisco Blogs","sourceLabel":"Cisco-on-Cisco vendor case study with self-reported infrastructure metrics","sourceUrl":"https://blogs.cisco.com/cisco-on-cisco/how-cisco-built-an-enterprise-ai-factory","evidenceClass":"vendor-claim","outcomeClass":"effective","topics":["infrastructure","developers-agents","data-security","operating-model"],"finding":"Cisco IT describes moving from project-specific GPU silos to a pooled multi-tenant AI factory after MyAgent rollout to 90,000 employees and Cisco IQ production hosting. Self-reported results: GPU utilization from ~15% to ~50% (peaks ~80%); active GPUs from ~250 to 1,568 without new real estate via rear-door heat exchangers; preference for ~80% steady-state inference on-prem; 99.99%+ SLA for MyAgent and Cisco IQ. Stack cited: Silicon One fabric, UCS-X with NVIDIA/AMD GPUs, Cisco AI Defense runtime guardrails, planned Hypershield segmentation, Live Protect on Nexus, Splunk Observability/Cloud/ITSI with Agent Observability and Tokenomics still expanding.","sledRelevance":"Relevant as an architecture pattern for SLED agencies and research campuses building on-prem/hybrid AI factories on Cisco networking and Splunk observability, not as a measured SLED deployment. Transferability requires matching power/cooling, multi-tenancy, FedRAMP/state privacy constraints and independent verification of utilization claims.","evidence":"Vendor-reported metrics and architecture narrative from Cisco IT authors; no independent audit, baseline methodology detail or third-party measurement of GPU utilization or SLA.","architectureImplications":"Interpretation: pooled high-bandwidth fabric, heterogeneous UCS GPUs, rack-level liquid assist and embedded AI Defense/Live Protect are the control points Cisco IT credits for utilization and security; SLED designs should validate Silicon One/Nexus and Splunk telemetry paths against local standards.","governanceImplications":"Interpretation: token visibility and closed-loop routing are described as in progress, so cost governance is incomplete even in this Cisco-on-Cisco story; agencies should require token/cost attribution before scaling agents.","securityPrivacyImplications":"Cisco embeds AI Defense for prompt/response inspection and plans Hypershield segmentation plus Live Protect virtual patching; claims are vendor architecture, not validated against a SLED threat model.","caveats":"Self-reported Cisco IT results; power, cooling and GPU supply assumptions may not hold in typical state data centers; Agent Observability/Tokenomics not finished.","streamIds":["cisco"],"roles":{"sales":"Sales framing: lead with infrastructure economics, not model demos. Cisco IT claims pooling and density lifted GPU use from about 15% to about 50% while hosting MyAgent and Cisco IQ. Ask which SLED workloads are stranded on dedicated clusters and whether on-prem inference share is a goal. Bound value to a measured utilization pilot; do not promise six times GPU growth or 99.99% SLA without local baselines and cooling proof.","engineering":"Engineering framing: treat the AI factory as a fabric-plus-security design. Validate Silicon One or Nexus pooling, UCS-X heterogeneous GPUs, rear-door cooling fit, AI Defense inline path, and Splunk metrics, logs, and ITSI before copying Cisco IT ratios. Constraints include power density, multi-tenant isolation, and unfinished Tokenomics. Proof of value is utilization, packet-drop, and guardrail metrics on one named workload class with a documented baseline.","delivery":"Delivery framing: stand up a shared AI landing zone with clear platform ownership. Dependencies include cooling plant, GPU supply, identity for agents, and Splunk observability. Skills span network fabric, MLOps, and SOC. Acceptance should require utilization and SLA evidence on a named workload; risks include stranded capacity returning if tenancy rules fail and unverified vendor metrics guiding bad buys. Label recommendations as interpretation and state limited applicability for SLED."},"roleDetails":{"sales":{"customerProblem":"Agencies and campuses need to scale agents without stranded GPUs, runaway cloud token spend or bolted-on security.","stakeholders":["CIO","CTO/infrastructure","CISO","FinOps/budget","Research computing lead"],"discoveryQuestions":["What is current GPU utilization and how siloed are clusters?","Which agent workloads must stay on-prem for sovereignty?","Do you already own Cisco networking, AI Defense and Splunk?"],"valueHypothesis":"Pooling fabric and denser cooling may raise GPU use and keep inference on-prem; Cisco’s 15%→50% figure is self-reported, not a SLED guarantee.","potentialEngagement":"A 6–8 week AI landing-zone assessment measuring baseline utilization, cooling headroom and security control gaps on one production-like workload.","unsupportedClaims":["Guaranteed tripling of GPU utilization","99.99% SLA in customer environments","Universal fit of rear-door heat exchangers"]},"engineering":{"fit":"Conceptual fit for Cisco-centric estates building multi-tenant AI factories; not a drop-in for non-Cisco fabrics.","architectureIntegration":"Pooled Silicon One/Nexus fabric, UCS-X NVIDIA/AMD GPUs, AI Defense runtime, planned Hypershield, Live Protect on Nexus, Splunk Observability/Cloud/ITSI.","prerequisites":["Sufficient power/cooling for denser racks","Shared scheduling/tenancy model","Identity and network path for AI Defense","Splunk metrics/logs entitlements"],"constraints":["Vendor-reported metrics only","Tokenomics/Agent Observability still expanding","Thermal plant may block density gains","No FedRAMP mapping in this post"],"security":"AI Defense prompt/response guardrails plus planned microsegmentation and Live Protect between patches; validate against agency ATO boundary.","proofOfValue":["Measure GPU utilization before/after pooling on one workload","Confirm zero GPU packet-drop stalls under load","Verify AI Defense blocks on a test prompt set","Track on-prem vs cloud inference cost share"]},"delivery":{"work":"Design and migrate from project GPU silos to a shared AI factory with security and observability wired in.","dependencies":["Facilities cooling upgrades","GPU procurement","Network fabric change windows","SOC playbooks for AI Defense alerts"],"ownership":"Joint platform ownership: infrastructure ops owns fabric/compute; security owns AI Defense/Hypershield; FinOps owns token attribution.","skillsAdoption":"Need fabric engineers, GPU/MLOps operators and security engineers fluent in agent guardrails; train app teams on shared tenancy rules.","governanceCheckpoints":["Architecture review of multi-tenant isolation","Security ATO delta for AI Defense path","FinOps gate before broad MyAgent-like rollout"],"acceptanceCriteria":["Documented utilization lift on named workload","Agreed SLA and on-call model","Token/cost dashboard for pilot agents","Written rollback to dedicated clusters"],"risks":["Self-reported Cisco metrics may not reproduce","Cooling project slips block density","Multi-tenant noisy-neighbor incidents","Unfinished Tokenomics leaves cost blind spots"]}},"retrievedAt":"2026-11-07T04:08:40Z","enrichedAt":"2026-11-07T04:14:10Z","enrichmentBasis":"retrieved source","accessibilityWorkforceImplications":"MyAgent at workforce scale implies change management and accessibility of agent interfaces for 90k employees; no ADA or public-sector workforce evidence is provided.","procurementImplications":"Stack spans compute, networking, AI Defense, Live Protect and Splunk entitlements; SLED buyers should map to existing Cisco/Splunk contracts rather than assume a single SKU AI factory.","operatingModelImplications":"Shifts from project-owned GPU islands to shared multi-tenant ops with enterprise SLA; requires FinOps for tokens and platform ownership across network, security and AI teams.","sourceVerification":{"responseId":"curl-cisco-ai-factory-20261106","openedUrl":"https://blogs.cisco.com/cisco-on-cisco/how-cisco-built-an-enterprise-ai-factory","referenceExcerpt":"Increased average GPU utilization from 15% to ~50%, with peak loads reaching 80%.","promptVersion":"sled-research-v4.0","model":null,"basis":"agent-reported inspection"}}}]}