Series: Getting Started & Tenant Foundations
Nothing about Fabric erases the years you already spent in Power BI Premium or Azure Synapse Analytics — it mostly renames furniture you already know how to use.
Premium becomes Fabric capacity.
Every Power BI Premium P SKU maps directly to a Fabric F SKU by capacity units — a P1 becomes an F64, a P3 an F256, up to P5/F1024. Microsoft is retiring P SKUs at the end of each term, with a 30-day grace period before throttling and a hard cutoff at day 91, so this isn’t optional homework. Mark that end-of-term date on your calendar now, because the throttling period is exactly when reports start refreshing late and nobody can figure out why.
Same content, new billing model.
In a same-region move, your reports, dashboards, semantic models, and refresh schedules land on the new F SKU with essentially zero downtime. What changes is flexibility: F SKUs bill per second, support pause-and-resume overnight, and include Power BI Embedded at every size instead of a separate EM/A SKU. That per-second billing is the part worth paying attention to, since a capacity you remember to pause overnight or on weekends can meaningfully cut your bill without touching a single report.
Synapse splits into Fabric experiences.
Azure Synapse Analytics bundled Spark pools, dedicated SQL pools, and pipelines into one service. Fabric separates those capabilities: Spark becomes Data Engineering, dedicated SQL pools become the Fabric Data Warehouse, and Synapse Pipelines becomes Data Factory in Fabric. Functionally you’re doing the same work, just under new names and inside a single workspace instead of a separate Synapse studio.
Warehouse migration needs a look, not a copy-paste.
Moving a Synapse dedicated SQL pool into a Fabric Warehouse only carries over raw data and metadata. Indexes don’t transfer, since Fabric tunes performance automatically, and a few data types like money map to different equivalents, so budget time to check your T-SQL. Treat this as a proper migration project with testing, not a weekend lift-and-shift — the queries that ran fine on Synapse deserve a second look before you trust them in production.
One capacity pool now, not two.
Premium capacity used to run mostly Power BI, while Synapse compute lived separately. A single Fabric capacity now shares its capacity units across every workload — Power BI, Data Engineering, Data Warehouse, Data Factory — all drawing from the same pool. That’s convenient for consolidated billing, but it also means a runaway Spark job can now compete for the same resources your executive dashboard depends on, so keep an eye on utilization as you add workloads.
This is orientation, not a choice.
You’re not being asked to pick Fabric over Premium or Synapse going forward; for existing customers, that decision is already made. Consider this a map so the terms you already trust carry over cleanly. Spend your energy on understanding the new vocabulary rather than relitigating a decision that’s already settled.
Where this leaves you: nothing new to decide, just vocabulary to relearn. Capacity, workspace, SKU sizing — the concepts survive intact. They’ve just moved in together.
To learn more: Power BI Premium to Microsoft Fabric Migration FAQ · Migrate workspaces from Power BI Premium capacity to Fabric capacity · Azure Synapse Analytics dedicated SQL pools to Fabric Data Warehouse migration · Compare Fabric and Azure Synapse Spark · Microsoft Fabric Capacity Metrics app (all Microsoft Learn)
This post is part of the Fabric & Power BI Essentials series — see the full series intro and article list there.
← Previous: What Is Microsoft Fabric? A Plain-English Overview for IT and Business Leaders | Next →: Setting Up Your First Fabric Tenant: Admin Portal Walkthrough
