Event streaming + KQL analytics + Real-Time Dashboards + Data Activator reactive alerts. Integrated with Fabric Warehouse + Power BI.
Last updated by Errin O'Connor, Founder & Chief AI Architect, EPC Group
Fabric Real-Time Intelligence: event streams + KQL Database + Real-Time Dashboards + Data Activator + Eventhouse. KQL (Kusto Query Language) used across Fabric + ADX + Sentinel + Monitor + Defender. Data Activator: reactive alerts triggering Power Automate + Teams + email + Power Apps + APIs. Six use cases: IoT telemetry, app observability, business KPIs, fraud/security, customer experience, supply chain. Uses Fabric F-SKU capacity: F64 (Microsoft list ~$5,003/month) supports 1M events/hr; F128 for 10M; F256+ for higher. Integration with OneLake + Warehouse + Power BI + Data Factory + Notebooks + Copilot for Fabric. EPC Group engagements: Single Real-Time Solution, Enterprise Program and IoT Program — each a fixed fee set at scoping.
Microsoft Fabric Real-Time Intelligence is the event streaming + analytics module within Fabric, providing: (1) Event streams — ingest events from IoT, Kafka, Event Hubs, custom APIs. (2) KQL Database — Kusto Query Language database optimized for time-series + telemetry data. (3) Real-Time Dashboards — live dashboards on streaming data. (4) Data Activator — reactive alerts + workflows on real-time signals. (5) Eventhouse — new fabric-native event store combining KQL DB + streaming. (6) Integration — with Fabric Data Warehouse + Data Lakehouse + Power BI. Replaces separate Azure Stream Analytics + Azure Data Explorer for Fabric customers.
Kusto Query Language (KQL) is Microsoft's query language for log + telemetry + time-series data. Used across: (1) Fabric KQL Database + Real-Time Intelligence. (2) Azure Data Explorer (ADX). (3) Azure Sentinel (SIEM). (4) Azure Monitor Logs. (5) Microsoft Defender + Purview. KQL syntax is designed for large-volume analytics with time-series operators, aggregations, joins, machine learning functions. Example: `Events | where Timestamp > ago(1h) | summarize count() by bin(Timestamp, 5m), EventType`. Learning curve: 20-40 hours for SQL developers, 40-80 hours for non-analytical users. Widely used across security + operations teams.
Data Activator is Fabric's reactive workflow engine — triggers actions when data conditions are met. Capabilities: (1) Real-time monitoring of streaming data or scheduled queries. (2) Threshold-based triggers (e.g., temperature > 80, revenue drops > 10%). (3) Action targets: Power Automate flows, Teams messages, email, Power Apps, custom APIs. (4) Multi-condition rules with logical operators. (5) No-code visual designer. Use cases: (1) IoT anomaly detection + alerts. (2) Business KPI threshold breach alerts. (3) Customer support ticket routing. (4) Fraud detection. (5) Supply chain exception handling. (6) SLA monitoring + escalation.
Six enterprise Fabric Real-Time Intelligence use cases: (1) IoT + industrial telemetry — equipment monitoring, anomaly detection, predictive maintenance. (2) Application observability — application logs + metrics + traces analysis. (3) Business real-time KPIs — hourly revenue, order volume, inventory, live executive dashboards. (4) Fraud + security — real-time transaction analysis + threat detection. (5) Customer experience — real-time website + app analytics + personalization triggers. (6) Supply chain — shipment tracking, exception detection, delivery updates. Common ingestion sources: Kafka, Event Hubs, Azure IoT Hub, custom REST APIs, Fabric Eventstream connectors.
Fabric Real-Time Intelligence uses Fabric F-SKU capacity — no separate SKU. F-SKU capacity units are consumed by KQL queries + streaming ingestion + Data Activator triggers + dashboards. Typical enterprise cost model: (1) Base Fabric F64 (~$8,410/month) supports moderate real-time workload (~1M events/hour). (2) F128 ($16,790/month) for heavy telemetry (10M events/hour). (3) F256+ ($33,580+/month) for very high volume. Compared to standalone Azure Data Explorer + Stream Analytics: often 20-40% cheaper for equivalent workload when consolidated in Fabric. Additional costs: OneLake storage $0.023/GB/month, no separate ingestion or query fees.
Integration points: (1) OneLake — KQL databases stored in OneLake, accessible from other Fabric workloads. (2) Fabric Data Warehouse — join streaming + warehouse data via SQL endpoint. (3) Power BI — build Real-Time Dashboards on Eventhouse data + DirectQuery to KQL. (4) Data Factory — pipelines can push data to KQL DB. (5) Data Science + Notebooks — analyze KQL data via Spark + Python. (6) Copilot for Fabric — natural-language KQL query generation. Real-Time Intelligence is deeply integrated — no separate ETL pipeline needed to bring data into Fabric analytics.
EPC Group Fabric Real-Time Intelligence portfolio: (1) Real-Time Discovery + Design (4 weeks) — use case validation + ingestion architecture + F-SKU sizing + fixed-fee proposal. (2) Single Real-Time Solution (8-12 weeks) — event ingestion + KQL DB + dashboards + Data Activator alerts. (3) Enterprise Real-Time Program (12-20 weeks) — multiple use cases + governance + team enablement. (4) IoT + Industrial Telemetry (10-16 weeks) — Azure IoT Hub + Fabric integration + predictive maintenance. (5) Ongoing Real-Time Engineering Retainer — monthly, scoped to the estate. Each engagement is a fixed fee quoted after a scoping call. All led by senior data engineering architect + KQL specialist.
4-week fixed fee: use case + architecture + F-SKU sizing. Call (888) 381-9725.
Monday-Friday, 8 AM - 7 PM CT
We respond to all inquiries within one business day
AI assistant — not human