Independent research institute for the physical limits of computeWorking research release · v11.10 · 2026.09
ResearchPublicationsDataPeopleInstituteSupport
ACTIVE PROJECTS / 2026.09

Research in progress.

Programmes define long-term questions. Projects show the work currently being developed, assembled or scoped, who is responsible, and what becomes public next.

Research programmesMethodologyActive projectsResearch agenda
CL-P01Method development

Compute Energy Measurement Protocol

How can physical GPU and host energy be turned into reproducible workload-level evidence without hiding allocation error?

LeadYan Zhang / AI & Data EngineeringProgrammesR01 · R03Current workCL-WP-005 v0.9 published as the methods + open-measurement synthesis release; primary telemetry campaign remains the next evidence gate.Published outputCL-WP-005 · The Physical Cost of Useful AI Work →
CL-P02Working research

Carbon of Compute

What is the full lifecycle carbon intensity of compute once semiconductor manufacturing, hardware, facility and operation share one boundary?

LeadRunwen Jin / Product & Carbon Systems; Xuguang Ma / synthesisProgrammesR02 · R05Current workLifecycle boundary, fab-to-compute allocation, public PCF benchmark bridge.Next outputWorking paper + open lifecycle schema.
CL-P03Working release published

Place Water Constraint Model

When does local water availability, seasonal stress or utility capacity become the binding limit on additional compute?

LeadCritical Load Research GroupProgrammeR04Current workSite water demand, basin context, threshold logic and evidence requirements.Current releaseCL-WP-004 · Peak Water, Not Average Water · v1.0; independent hydrology review remains the next validation gate.
CL-P04Evidence assembly

CL-001 Place Assessment

Can one place assessment carry grid, water, lifecycle carbon, land, permitting and community evidence into a defensible capacity and decision record?

Project leadXuguang MaData systemYan ZhangPolicy adviceDr. Neo Lin · Yumeng LiuNext outputBaseline assessment. No Critical Load rating has been issued.
CL-P05Working research

Constraint-aware Compute Routing

How should flexible workloads move across time and place after latency, sovereignty, grid, water, reliability and cost constraints are applied?

LeadYan Zhang / AI & Data EngineeringProgrammeR06Current workFeasible-set logic, multi-objective comparison and decision-record design.Next outputResearch note + reproducible decision examples.
CL-P06Scoping

Open Evidence Registry

How can every published Critical Load result preserve sources, assumptions, uncertainty, method version and correction history?

LeadResearch & Data teamProgrammesCross-programmeCurrent workEvidence classes, source IDs, publication registry and correction fields.Next outputRegistry specification; public code repository not yet released.