ATX AnalyticsTable reports on SQL Server where it already runs — on-premises, often next to the analyst — and keeps the whole chain local. Windows integrated sign-in means the query runs as the person at the keyboard; the result lands in a DuckDB database in the project; reports and scheduled PDFs are produced on the same machine or your own server.
The Microsoft SQL Server connection form in ATX AnalyticsTable 1.7, filled in with the example values below. Secrets go into the encrypted vault, never into the project.
Read access for an AD group (as a DBA): CREATE LOGIN [CORP\Risk-Analysts] FROM WINDOWS; USE RiskDW; CREATE USER [CORP\Risk-Analysts] FOR LOGIN [CORP\Risk-Analysts]; ALTER ROLE db_datareader ADD MEMBER [CORP\Risk-Analysts];
Reporting query — client tax IDs hashed in SQL Server: SELECT p.trade_date, p.desk, CONVERT(char(64), HASHBYTES('SHA2_256', UPPER(c.client_tax_id)), 2) AS client_key, SUM(p.exposure_eur) AS exposure_eur FROM dbo.positions AS p JOIN dbo.clients AS c ON c.client_id = p.client_id WHERE p.trade_date >= DATEADD(day, -30, CAST(GETDATE() AS date)) GROUP BY p.trade_date, p.desk, CONVERT(char(64), HASHBYTES('SHA2_256', UPPER(c.client_tax_id)), 2) ORDER BY p.trade_date;
Then reshape locally — a Run SQL step on the project's DuckDB: -- "input" is the table the previous pipeline step produced (the import above). -- This runs on your machine, in the project's DuckDB, not on the database. SELECT region, SUM(revenue) AS revenue, COUNT(DISTINCT customer_key) AS customers FROM input GROUP BY region ORDER BY revenue DESC; — Column names follow the query above; on the local engine the pipeline's current table is registered as input before the script runs.
SQL Server Dynamic Data Masking, row-level security and column permissions apply to the login that runs the query. Windows integrated security already runs as the person at the keyboard; on Azure SQL, the Azure AD modes with on-behalf-of do the same for a server installation.
SQL Server on your network plus AnalyticsTable on your machine means no internet is needed at all. After the import, the data and every report over it also work with the database unreachable — on a laptop away from the office, for example. The licence validates offline.
For many analyst-owned reports, yes: queries, pivots, dashboards, PDF reports and scheduled e-mail, built in a grid rather than a report designer. It runs on the analyst's desktop or as AnalyticsTable Server on your own Linux servers.
With Windows integrated security there is none. Other methods keep their secret in an encrypted vault, never in the project file, and connection exports strip secrets.
Yes, for the login that runs the query — with Windows sign-in that is the analyst. AnalyticsTable's own masking then protects the extract that leaves the machine.
This site needs JavaScript for the full experience. Every page above is reachable without it.