Skip to content

Model

The developer's by-kind index: where each CORA aggregate's I-TOMCAT content lives, the SLS 2.0 control-stack seam this exercise draws, and the record of what is deliberately deferred. Design-phase scaffold.

For the aggregate shapes see the architecture model and the per-BC modules.

Aggregate (BC) Where at I-TOMCAT
Asset (Equipment) the stage pages: Source, Sample, Detector
Capability, Method (Recipe) Techniques
Enclosure (Enclosure) the index
Zone, Conduit, Policy (Trust); Actor (Access) Governance
Procedure, Recipe, Caution, Supply, Subject, Run, Campaign, Dataset, Decision deferred (design-phase; see below)

The seam: what CORA would replace vs drive through

I-TOMCAT is the fleet's view of an SLS 2.0 beamline, so the control-stack boundary matters more than usual. SLS is an EPICS facility with the BEC (Beamline and Experiment Control) scan layer over ophyd introduced for SLS 2.0.

Layer SLS tool CORA seam
Control (floor) EPICS IOCs drive through (never replaced; the floor CORA actuates and observes)
Scan / orchestration (edge) BEC over ophyd replace (CORA's edge replaces BEC's scan/experiment steering; the BEC-shares-ophyd nuance keeps a drive-through reading open, SEAM-1)
Detector / capture the camera streaming + HDF5 writer chain drive through / observe (specialized capture CORA observes)
Data-of-record SciCat + the Ra/SLURM Fiji reconstruction pipeline replace / invert source-of-truth (CORA owns its own Dataset; SciCat is a fact a future integration reads, not CORA's record)

This is the standard CORA lens (EPICS is the floor, the facility's scan/data software is named only to draw the boundary). The single most consequential call, BEC replace vs drive-through, is carried as SEAM-1 on Open questions because BEC adopts the same ophyd device model CORA's edge would.

What is deliberately not here yet

  • Integration scenarios. No test_i_tomcat_*.py registers I-TOMCAT Assets into the event store. Scenario code is where Assets become real, and hard-registering a modelling-exercise beamline with unconfirmed facts would commit speculative structure. It lands if the deployment firms toward a real connection.
  • Vendor Models. No catalog Model is bound. The "(target)" models in the descriptor are open questions, not bindings, because they are read from public pages and not staff-confirmed.
  • Operations and experiment views. A runbook and live experiment view for a beamline CORA has not connected to would be invention; see the note on the index.