Stream Analytics Job — workload reference

Family: Data Engineering Loom slug: stream-analytics-job Editor file: apps/fiab-console/lib/editors/stream-analytics-editor.tsx BFF routes: app/api/items/stream-analytics-job/** Bicep module: platform/fiab/bicep/modules/landing-zone/stream-analytics.bicep (added in the Data Engineering sweep, 2026-05-27)

Purpose

Authoring + lifecycle control for Azure Stream Analytics (ASA) jobs. Replaces the deprecated U-SQL / ADLA editor (the underlying service is retired). The editor lists every ASA job in the configured RG, shows state (Stopped / Starting / Started / Stopping / Failed), exposes the SAQL query (Stream Analytics Query Language — SQL over time-windowed streams), and supports Save query / Start / Stop via ARM REST.

Fabric-parity gap

Stream Analytics is not a Fabric native item — Fabric's equivalent is Eventstream (which Loom ships as a separate editor, eventstream). ASA exists in this catalog for the streaming use-cases that need:

• Native joins across Event Hubs + Blob/SQL reference data
• ASA's deterministic windowing semantics
• Customers who already run ASA in production

Capability	Loom state
List jobs	Shipped — ARM streamingjobs?api-version=2020-03-01
Job detail (inputs/outputs/transformation)	Shipped — $expand=inputs,outputs,transformation,functions
Edit + save SAQL query	Shipped — PUT streamingjobs/{name}/transformations/{xName}
Start / Stop	Shipped — POST /start and /stop
Create new input / output	Gated — surfaced as MessageBar pointing at portal until v2
Monitoring metrics (SU%, watermark)	Surfaced from lastOutputEventTime + streamingUnits; deep Azure Monitor link is roadmap

Real backend it calls

• @/lib/azure/stream-analytics-client.ts (added in this sweep) — ARM REST against Microsoft.StreamAnalytics/streamingjobs using ChainedTokenCredential(UAMI, default).
• 501 gating: if LOOM_ASA_RG is unset, the BFF returns {ok: false, hint: …} with the env var + bicep module the operator must wire. The editor renders that as a Fluent MessageBar.

Sample usage

1. Deploy the optional ASA module with enableStreamAnalytics=true.
2. Open /items/stream-analytics-job.
3. Pick a job from the left list.
4. Edit the SAQL query (e.g. tumbling-window average per device).
5. Save to PUT the transformation.
6. Start to begin processing.

Bicep + env vars

Env	Purpose	Bicep module
LOOM_ASA_RG	RG containing ASA jobs	landing-zone/stream-analytics.bicep
LOOM_ASA_SUB	Subscription (defaults to LOOM_SUBSCRIPTION_ID)	same
LOOM_UAMI_CLIENT_ID	Console UAMI	identity.bicep

The UAMI needs Stream Analytics Contributor on the configured RG; the bicep module issues that role assignment automatically.