1. IoT Data Analytics
  2. Features
  3. Execution provenance
Feature · Execution provenance

Execution provenance

Trace each run, numbered step and stream state transition.

OverviewRuns needing reconciliation
IoT Data Analytics · Execution provenance

Runs needing reconciliation

Operation runs

New
Run numberOperation keySource revisionStatusFailure reasonSource
OR-68863OK-318450Reconciliation requiredCut to lengthSS-55954
OR-46504OK-790940Reconciliation requiredWeld frameSS-33497
OR-72272OK-207570Reconciliation requiredDrill and tapSS-61889
OR-96681OK-621370Reconciliation requiredDeburrSS-21673
OR-93583OK-280620Reconciliation requiredAssembleSS-24987
6 of 6 records
OK-318Operation runs · OR-68863
Reconciliation required
Source revision450
Failure reasonCut to length
SourceSS-55954
Example screen with sample data

What you can do

Trace and correct an execution

Operation runs hold the operation key, source revision, input hash, status, result reference and failure reason. Operation steps record each numbered action with input and output references and any external receipt evidence. State history attaches a status transition, actor, time, reason and operation key to the stream.

How the work is organized

  • Run. The Operations Analyst checks the operation key, frozen revision and result reference.
  • Steps. Each numbered action records its status and input and output references.
  • History. The Stream Owner compares the current status with recorded transitions.

This trail helps the reviewer separate an attempted action from a confirmed effect. It also connects a metric result to the exact run that produced it, so a replacement calculation does not erase the earlier answer.

What to agree during setup

Agree how repeated operation keys are handled and when a run requires reconciliation. If an external step returns an unknown result, inspect the receipt before retrying. Define who may correct an executed revision and how the successor is identified.

Rehearse a repeated operation key and a run whose effect needs reconciliation. The Operations Analyst should distinguish the existing run from a new attempt and inspect any step receipt. The Stream Owner checks the status transition and source revision before making a correction.

Check the handoff on an actual case

Suppose a run shows a changed stream but its final read-back failed. The analyst inspects the numbered steps, compares the result reference to the stream and determines whether the effect exists. They keep the run observable until reconciliation resolves it. If a correction is needed, the owner increments the stream revision; the earlier run and history remain intact.

Ask the Operations Analyst to trace the correction from the source revision to its run, numbered steps and state history. If a metric result is involved, compare the original run and result with the successor rather than overwriting the first calculation.

For a rehearsal, select one successful run and one run needing reconciliation. Have the analyst identify the input hash, source revision and result reference in each. The distinction should remain clear when the stream has already moved to a later revision.

What your team sees

  • Runs needing reconciliation — Runs whose outcome needs an analyst to investigate.
  • Stream status board — Current stream states for comparison with recorded history.

Follow the work through

See the related process for the handoff and the data model for the connected records.

What it keeps track of

Data model
Operation runs
Workflow execution runs with frozen revision, input hash and result identity.
Operation steps
Numbered steps of an operation run with input/output and receipt state.
State history
Immutable lifecycle transition history for each source.
Sensor streams
Measured web properties with lifecycle status and observed outcome fields.
Metric results
Calculated metric results persisted once per approved definition and window.

Who uses it

Permissions
Operations Analyst
, Stream Owner, Measurement Steward.

Set up IoT Data Analytics your way.

Start free, then use Proto to change fields, steps and approvals before your team goes live.

How to get started