{"resourceId":"rit-tigris-research-computing-20260925","versions":[{"version":"external-0eda27531243145ce9fcf56c3afec4cdb37a9dbe84ec2e22deb1ad0972864a06","resource":{"id":"rit-tigris-research-computing-20260925","title":"RIT TIGRIS: campus HPC cuts average job wait from ~6 hours on SPORC to ~18 minutes; TOP500 #302 / Green500 #82","organization":"Rochester Institute of Technology (RIT) Research Computing","sector":"Higher education research computing","geography":"United States (Rochester, NY)","publishedAt":"September 25, 2026","publicationDate":"2026-09-25","eventDate":"2026-09-25","sourceName":"RIT News","sourceLabel":"University news (RIT)","sourceUrl":"https://www.rit.edu/news/rit-research-computing-earns-international-recognition-new-supercomputer","evidenceClass":"independent-reporting","outcomeClass":"effective","topics":["infrastructure","operating-model","developers-agents"],"finding":"RIT reports that TIGRIS, launched summer 2026, delivers about 3× CPU and 7× GPU performance versus prior cluster SPORC, ranks No. 302 on TOP500 and No. 82 on Green500 (No. 18 / No. 15 among U.S. academic systems on those lists), and that average job wait fell from about six hours on SPORC to about 18 minutes on TIGRIS. Research Computing cites support for 830+ scholarly works since 2016 and shared services across faculty and 1,600+ students.","sledRelevance":"Campus research computing leaders evaluating AI/HPC capacity upgrades need measured queue-time and ranking outcomes, not only GPU counts, when justifying shared clusters for AI-assisted discovery workloads.","evidence":"RIT News article dated September 25, 2026 opened via HTTPS; queue-wait, TOP500/Green500, and 3×/7× performance claims taken from page body.","architectureImplications":"Shared GH200-class campus cluster as university-wide batch resource alongside prior SPORC; Research Computing model bundles HPC, storage, training, workflow optimization, and grant facilitation rather than PI-owned islands.","governanceImplications":"Institutional ownership of fair-share scheduling and shared capacity implies research-admin policies for priority, training, and allocation—not only capital purchase of GPUs.","securityPrivacyImplications":"University news does not detail tenant isolation, research-data classification, or export controls; treat as capacity announcement requiring local security review before AI-agent or sensitive-data use.","caveats":"Self-reported institutional metrics; no independent replication of queue statistics; ranking snapshot dates not fully specified beyond list membership.","streamIds":["research"],"roles":{"sales":"Sales framing: Research computing and VPR buyers want shared AI/HPC capacity that actually shortens time-to-result, not another underused GPU island. RIT’s TIGRIS story supports a pitch around measured queue reduction (~6 hours→~18 minutes versus SPORC), TOP500/Green500 placement, and a Research Computing service wrapper (storage, training, consultation) that has already co-supported 830+ scholarly works. Keep claims bounded to campus batch acceleration and energy-efficiency ranking—not autonomous discovery outcomes.","engineering":"Engineering framing: Plan a university-wide Slurm-style GH200/ARM cluster that coexists with legacy x86 GPU nodes, with fair-share scheduling and Research Computing as the ops owner. Integration work is storage paths, software stacks for AI and traditional HPC, queue telemetry for wait-time SLOs, and training pipelines so PIs do not buy private boxes. Proof of value is queue-start distributions and utilization, not vendor peak FLOPS alone.","delivery":"Delivery framing: Stand up shared cluster access, migrate representative AI and simulation workloads from legacy queues, publish wait-time baselines, and train users on fair-share and right-sizing. Acceptance should include documented average/percentile queue starts versus the prior system, Green500/TOP500 or equivalent efficiency reporting if claimed, and a service catalog (storage, consult, training) with named owners—not a bare metal drop."},"roleDetails":{"sales":{"customerProblem":"Campus researchers lose days in HPC queues while AI and simulation demand outstrips legacy shared clusters.","stakeholders":["VPR/research administration","Research computing director","College deans","Faculty power users","CIO/facilities for power/cooling"],"discoveryQuestions":["What is current average/percentile queue wait by partition?","Which AI vs simulation workloads drive GPU demand?","Is Research Computing funded as a shared service or recharge-only?","What TOP500/Green500 or efficiency goals matter to leadership?"],"valueHypothesis":"A shared next-gen GPU cluster plus Research Computing services can cut queue waits by an order of magnitude versus a saturated legacy cluster, if utilization and training keep the system shared.","potentialEngagement":"Queue-telemetry baseline and 90-day migration pilot for 3–5 AI/HPC groups with wait-time and utilization SLOs.","unsupportedClaims":["Guaranteed discovery breakthroughs","Zero cooling/power constraint","Replacement of all PI-owned GPUs","Autonomous lab outcomes"]},"engineering":{"fit":"University-wide ARM/GH200 HPC cluster coexisting with legacy x86 GPU partitions under Research Computing ops.","architectureIntegration":"Integrate Slurm fair-share, parallel filesystems, AI software modules, identity, and queue telemetry dashboards; keep SPORC-class systems for transitional workloads.","prerequisites":["Power/cooling headroom","Research storage capacity","Identity and allocation policy","User training program","Baseline queue metrics"],"constraints":["Institutional self-report of waits","Single-campus deployment","Ranking lists are snapshots","Does not specify agent safety controls"],"security":"Define data classification, multi-tenant isolation, and export-control paths before hosting sensitive or regulated research on the shared AI partition.","proofOfValue":["Average/percentile job start wait vs prior cluster","GPU/CPU utilization","User-reported time-to-result on named workloads","Published efficiency ranking if claimed"]},"delivery":{"work":"Commission cluster, migrate pilot workloads, publish wait-time SLOs, and stand up Research Computing consult/training.","dependencies":["Facilities power/cooling","Network/storage","Identity/IAM","PI pilot groups","Communications for cutover"],"ownership":"Research Computing owns platform ops and queue policy; PIs own scientific methods; VPR owns allocation governance.","skillsAdoption":"Train users on fair-share, right-sizing GPUs, and module stacks; train consultants to read queue telemetry.","governanceCheckpoints":["Allocation policy sign-off","Security/data-class review","Pilot SLO review at 30/90 days"],"acceptanceCriteria":["Documented wait-time improvement vs baseline","Service catalog live","Pilot groups productive on TIGRIS-class partition"],"risks":["Queue metrics regress under load","PI shadow IT persists","Power/cooling limits throttle GPUs","Overclaiming rankings as scientific quality"]}},"retrievedAt":"2026-09-26T05:20:56-05:00","enrichedAt":"2026-09-26T05:21:10-05:00","enrichmentBasis":"retrieved source","accessibilityWorkforceImplications":"Shared cluster plus Research Computing training expands access for students and faculty who cannot buy private GPUs; still depends on fair-share literacy.","procurementImplications":"Justify shared cluster CapEx/OpEx with wait-time SLOs and service wrapper, not GPU brochure FLOPS alone; include Green500-style efficiency in RFP scoring if energy goals matter.","operatingModelImplications":"Research Computing as university-wide service (HPC, storage, consult, grant help) is the operating model that converts hardware into reduced wait times.","sourceVerification":{"openedUrl":"https://www.rit.edu/news/rit-research-computing-earns-international-recognition-new-supercomputer","referenceExcerpt":"Computing jobs that waited an average of six hours to begin processing on SPORC now start in about 18 minutes on TIGRIS.","promptVersion":"sled-research-v4.0","model":null,"basis":"agent-reported inspection"}}}]}