Cloud Pro: Azure Analytics: Choose, Compare and Launch

What's inside

1. Level 1: The Azure Analytics Map, Old and New NamesFree
   Put every Azure analytics and data service on one page, sorted into seven groups (warehouses and lakehouses, query engines, data integration and pipelines, streaming and messaging, big data processing, catalogues and governance, business intelligence), each with its current and former names and its status: generally available, maintenance mode or retired.
2. Level 2: The Standard Analytics ProfileFree
   Build the reusable profile that every later choice is scored against: sources and volume, batch or streaming, how fresh insight must be, query patterns and concurrency, who the users are, open formats, governance and lineage, residency, team skills and cost model.
3. Level 3: Fabric Foundations: OneLake and Capacity
   Learn how Microsoft Fabric is put together: one SaaS tenant, OneLake as the single Delta-based store, shortcuts, and the F-SKU capacity model with capacity units, smoothing, bursting and throttling that decides both performance and the bill.
4. Level 4: Fabric Data Factory: Pipelines and Dataflows
   Cover data integration inside Fabric with pipelines, Dataflow Gen2, copy jobs and mirroring: what each one is for, how they use capacity, and where they differ from the stand-alone Azure Data Factory.
5. Level 5: Fabric Data Engineering: Lakehouse and Spark
   Work with Fabric's lakehouse, notebooks, Spark job definitions and environments: how Spark sessions use capacity, starter pools versus custom pools, and V-Order and Delta table maintenance.
6. Level 6: Fabric Data Warehouse and SQL Endpoints
   Understand the Fabric Data Warehouse: full T-SQL reads and writes over Delta in OneLake, how it differs from a lakehouse's read-only SQL analytics endpoint, and what its T-SQL coverage, workload management and limits mean for migrations from Synapse or SQL Server.
7. Level 7: Fabric Real-Time Intelligence
   Learn Fabric's streaming stack: Eventstreams for ingestion and routing, Eventhouse and KQL databases for fast analytics over time-series data, Real-Time Dashboards and Activator for alerts, plus how this relates to stand-alone Event Hubs and Data Explorer.
8. Level 8: Power BI on Fabric: Semantic Models and Direct Lake
   Place Power BI inside Fabric: Import, DirectQuery and Direct Lake storage modes, semantic model design, Pro and Premium Per User licences compared with F-SKU capacity, and the fallback and refresh behaviour that shapes cost and speed.
9. Level 9: Azure Databricks: Platform and Compute
   Learn the Azure Databricks platform: the split between control plane and compute plane, classic and serverless compute, all-purpose versus jobs compute, Photon, DBU-based pricing, and what you operate compared with what Databricks and Microsoft operate.
10. Level 10: Databricks Unity Catalog and SQL Warehouses
    Cover the governance and BI side of Databricks: Unity Catalog metastores, catalogues, external locations, lineage and Delta Sharing, plus serverless, pro and classic SQL warehouses for analyst concurrency.
11. Level 11: Azure Synapse Analytics and Its Future
    Assess Synapse as it stands today: dedicated and serverless SQL pools, Spark pools, Synapse pipelines and Data Explorer pools. It is supported but in maintenance mode, with Microsoft putting new investment into Fabric and offering migration tooling.
12. Level 12: Azure Data Factory: Integration at Scale
    Learn stand-alone Azure Data Factory: pipelines, activities, Azure, self-hosted and SSIS integration runtimes, mapping data flows, triggers and its pay-per-activity pricing, and how it sits alongside Fabric Data Factory.
13. Level 13: Azure Event Hubs: Streaming Ingestion
    Master Event Hubs as the streaming front door: partitions, consumer groups, throughput and processing units, Standard, Premium and Dedicated tiers, Kafka compatibility, Capture to ADLS and the limits that cap throughput.
14. Level 14: Azure Stream Analytics: SQL over Streams
    Learn Stream Analytics jobs: inputs, outputs, windowing functions, late-arriving and out-of-order events, streaming units and parallelism, and where a SQL-first managed stream processor is the right tool.
15. Level 15: Azure Data Explorer: Fast Time-Series Analytics
    Cover Azure Data Explorer (Kusto) clusters for log, telemetry and time-series analytics: ingestion methods, hot cache versus retention policies, KQL, cluster SKUs and autoscale, and how it relates to Fabric Eventhouse and to the free cluster.
16. Level 16: Azure HDInsight: Managed Open-Source Clusters
    Assess HDInsight, which runs managed open-source clusters (Spark, Hadoop, Kafka, HBase): its supported versions and dated end-of-support notices, the retirement of HDInsight on AKS, its always-on cluster cost model and its narrow remaining fit.
17. Level 17: Microsoft Purview: Catalogue and Governance
    Learn the data governance side of Microsoft Purview: the Data Map, scanning and classification, the Unified Catalog, lineage from Data Factory, Fabric and Databricks, and its pricing and limits.
18. Level 18: Lakehouse Architecture and Open Table Formats
    Compare Delta Lake and Apache Iceberg (with a note on Hudi) across Azure services: transaction logs, time travel, schema evolution, partitioning and liquid clustering, and how OneLake shortcuts, metadata conversion between formats, UniForm and Unity Catalog let engines share one copy of the data.
19. Level 19: Data Governance and Lineage End to End
    Design governance across a whole Azure estate: one owner per catalogue, end-to-end lineage across pipelines, Spark and BI, sensitivity labels, data products and data contracts, and where automatic lineage stops so you must fill gaps by hand.
20. Level 20: Security, Networking and Data Residency
    Apply security and residency to analytics services: Entra ID identities and managed identities, private endpoints and managed VNets, customer-managed keys, row- and column-level security, and the regional and data-boundary rules for where data and metadata are stored and processed.
21. Level 21: Streaming Design: Events, Windows and Guarantees
    Learn how to design a streaming path from start to finish on Azure. Decide on partitioning, event time versus arrival time, windowing, late and duplicate events, delivery guarantees and where processed streams land, then choose between Event Hubs with Stream Analytics, Fabric Real-Time Intelligence, Databricks Structured Streaming and Data Explorer.
22. Level 22: Migration Paths Between Azure Data Services
    Learn how to plan moves between Azure data services: Synapse to Fabric or Databricks, HDInsight to Databricks or Fabric, and Azure Data Factory to Fabric Data Factory. Cover what moves easily, what has to be rewritten and how to run the old and new systems side by side safely.
23. Level 23: Worked Cost Model: A Reporting Workload
    Build a dated cost model for a batch reporting workload: nightly loads, a modelled warehouse or lakehouse and Power BI users. Price it on Fabric capacity, on Databricks SQL and on Synapse, and see which levers change the bill.
24. Level 24: Worked Cost Model: A Streaming Workload
    Build a dated cost model for a continuous streaming workload covering ingestion throughput, always-on processing, hot storage and retention. Compare the options and find where streaming costs differ most from batch.
25. Level 25: Building a Weighted Selection Scorecard
    Turn the standard analytics profile into a weighted scorecard with criteria, weights, scoring rubrics and knock-out rules, so that service choices can be repeated and defended.
26. Level 26: Scoring Reference Scenarios on Azure
    Apply the scorecard to three or four contrasting reference scenarios, such as a mid-size BI estate, an IoT telemetry platform, a data science lakehouse and a regulated multi-region business, and reach a defended Azure recommendation for each.
27. Level 27: Scoring Azure Against Another Cloud
    Score the chosen Azure design against the nearest equivalent on another cloud, such as AWS or Google Cloud, and support every score with dated, sourced evidence for each service, limit and price.
28. Level 28: Final Challenge: Decision Workshop and Selection Record Exam
    Run a full decision workshop on a new brief, using the whole course: build the profile, shortlist and score services, model costs, compare against another cloud and sketch a launch plan. Finish with a one-page selection record you can defend.

Access

The first 2 levels are free with a free account. Every level, the podcast edition and the AI tutor come with All Access at £4.99/month or any Creator plan — see pricing.

Start free →