SSAS replacement · SSAS alternative
The SSAS replacement for a modern data warehouse
Your data moved to ClickHouse, Snowflake or BigQuery. Microsoft Analysis Services did not follow — and your analysts lost the Excel PivotTables they ran the business on. XLTable is an XMLA-compatible OLAP server that speaks Excel’s language on top of the warehouse you have now: the same PivotTables, hierarchies and MDX, with the data staying where it lives.
Why SSAS stops working after the warehouse migration
It cannot query the new warehouse live
Analysis Services is built around SQL Server. To use ClickHouse, Snowflake or BigQuery you copy the data into the cube and reprocess it on a schedule — a second copy of the warehouse, an ETL job to babysit, and numbers that are hours old.
BI dashboards do not replace PivotTables
A dashboard answers the questions someone predicted. Analysts need to drag a new dimension onto the rows and see the number now — so they export to Excel anyway, and governance is gone.
Keeping SSAS means keeping SQL Server
SSAS ships with SQL Server: per-core licensing, a Windows-only stack and a team to run processing — all for a cube that no longer holds the source of truth.
Replacing SSAS — with or without a warehouse
Your users keep their Excel PivotTables.
Your warehouse
Already on ClickHouse, Greenplum, BigQuery or another engine? XLTable connects to it. No data copies — your database does the math.
XLTable Engine — StarRocks, turnkey
No warehouse yet? We install StarRocks next to XLTable, set up scheduled loads from SQL Server, PostgreSQL and ERP, and materialized views for your cubes. Support is on us.
For your users, nothing changes
PivotTables connect over XMLA, just like to SSAS — same hierarchies, drill-down and slicers. No retraining.
Same XLTable either way: it turns PivotTable queries into SQL and the database does the math.
Talk to us about your SSAS migrationWhat stays the same
- • Excel connects natively over XMLA through the standard Analysis Services dialog — no add-in, no retraining.
- • Dimensions, hierarchies, measures, calculated fields, drillthrough and MDX work as your analysts expect.
- • Row-level security and Active Directory / LDAP — one governed model for everyone.
- • Report layouts survive: keep the cube, dimension and measure names and existing PivotTables keep their shape.
What changes
- • The cube is defined in SQL and kept in Git — no Visual Studio project, no DAX.
- • No cube to process: every PivotTable action becomes one SQL query in the warehouse — on your own warehouse, with no data copies.
- • Runs on Linux or Windows, on-premises or in your cloud, air-gapped if needed.
- • Two more ways to work with the same cubes: a pivot table in the browser, with reports shared by link, and AI assistants through the MCP server included with XLTable.
Check it in two minutes
Connect Excel to our demo server — the way you connect to SSAS
A public XLTable server with sample retail data. Open the Analysis Services dialog in Excel, enter the server address and the shared login, and pick the cube. It is the dialog your analysts already use for SSAS — no add-in, nothing to install.
Data → Get Data → From Database → From Analysis Services
The live demo: Excel, the browser and AI on the same cubes →- Server
- https://olap-demo.xltable.com/
- User
- demo
- Password
- demo
- Database
- demo
- Cube
- Demo Sales
SSAS vs XLTable
Condensed from the full feature comparison in the documentation.
| Feature | SSAS | XLTable |
|---|---|---|
| Data source | SQL Server first; other sources are imported into the cube | ClickHouse, Snowflake, BigQuery, Databricks, StarRocks, Trino, Greenplum, DuckDB — queried live; no warehouse yet: XLTable Engine |
| Copy of the data | Yes — the cube stores its own copy | None on your warehouse — every query is pushed down to it |
| Processing / refresh jobs | Required (full or incremental) | No cube processing — the numbers are as fresh as the warehouse; XLTable Engine loads data on a schedule |
| Cube definition | Visual Studio project | SQL files in Git |
| Excel PivotTables over XMLA | ✓ | ✓ |
| MDX, hierarchies, calculated fields | ✓ | ✓ |
| Row-level security, Active Directory / LDAP | ✓ | ✓ |
| Drillthrough to detail rows | ✓ | ✓ |
| Pivot table in the browser | — | ✓ the same cubes without Excel, reports shared by link |
| AI assistants (MCP) | — | ✓ MCP server included |
| DAX, Power BI live connection | ✓ | — |
| Platform | Windows + SQL Server | Linux or Windows, on-premises or your cloud, air-gapped |
| Licensing | Per SQL Server core | Per user; per server for large deployments |
Migration plan: cube by cube
SSAS and XLTable run side by side. You move one cube, compare the numbers and switch its users — then take the next one. SSAS keeps serving the cubes you have not moved yet.
01
Install XLTable next to SSAS
Install the service on a Linux or Windows server and add the connection to the warehouse — ClickHouse, Snowflake, BigQuery or any other supported engine. Most installations take under an hour. No warehouse yet? We set up XLTable Engine next to it.
02
Rebuild the first cube in SQL
Start with a cube people open every day. Generate the first draft from the warehouse tables in seconds, then port dimensions, measures and access rules. The definition is one SQL file kept in Git.
03
Compare the numbers
Open the same PivotTable against SSAS and against XLTable and compare the totals — before anyone is switched.
04
Reconnect Excel
Analysts change the server address in the Analysis Services connection. With the same cube, dimension and measure names their PivotTables keep their layout.
05
Take the next cube
Repeat for the other cubes, one at a time, in the order that suits the business.
06
Retire cube processing
When the last cube is moved, switch off SSAS processing. XLTable keeps no copy of its own and reads the warehouse directly; with XLTable Engine, the scheduled loads into StarRocks take the place of cube processing. Shared query caching and pre-aggregation-friendly SQL keep PivotTables fast on billions of rows.
What happens to MDX, roles and your Excel users
- MDX
- Excel keeps sending MDX; XLTable turns each query into SQL for the warehouse. Calculated members of the SSAS cube are rewritten as calculated fields — ordinary SQL expressions of your warehouse, stored in the cube file.
- Roles and row-level access
- An SSAS role becomes a role block in the cube file: the Active Directory groups it covers, the measures and dimensions they see, and the row filters. XLTable adds those filters to every SQL query it sends to the warehouse.
- Excel users
- Nothing changes: the same Analysis Services dialog, the same PivotTables, slicers and drill-down. Only the server address is new.
- How long it takes
- That depends on how many cubes you have and how much calculation logic they carry, so we give an estimate after looking at your cubes. The installation itself is quick: Lenta had XLTable deployed in one working day.
Proven in production
Retail chain Lenta runs Excel self-service for ~80 concurrent users on billions of rows in ClickHouse with XLTable — deployed in a single working day, on a perpetual license, with no extra database or BI servers.
Read the Lenta case study →Further reading
Ready to plan a migration?
Start on your laptop today with the free desktop edition — the cube files move unchanged to the server when you are ready — or talk to us about a pilot in your environment.