I would like to be able to run a "lineage/dependency" report in Sisense. Something that will tell me for all the cube/live data sources:
- Data source name
- Type (cube/live)
- Dates of the data model (created, modified, last built)
- Data sources being referenced and type (web service, database, etc.)
- Details of each data sources (SQL Server, servername, database name)
- Tables or objects being referenced by those data sources
It's important for managing the environment to know what we are using, or have an easy way to know what data sets/dashboards would break if a table's structure were changed in the database.
Data Source Dependency Report
5 Comments
dougnewton Since this is from 2024, I'm not sure if you're still in the Sisense-sphere, but I wanted to jump in regardless.
Totally hear you on this! It makes complete sense that this type of report/data should just be built natively and available in Sisense.
While we can’t speak for the product team’s roadmap, this is something we've heard a lot from other Sisense customers. Change impact is tricky, and knowing what will actually break before you touch something in the database is really valuable.
We built the Asset Auditor to address exactly what you're describing. It covers all live and Elasticube data models (not unbuilt, though) and surfaces the connection details, database and schema names, and the tables and columns being referenced.
From there, you can trace dependencies all the way through to the dashboards and widgets consuming that data, so if a table structure changes, you know exactly what's at risk before anything breaks. We also surface existing errors that are related to missing data models or missing tables/columns, so that you can get them cleaned up. And that is just the start!
Here is an example of one of the (many) widgets the Asset Auditor provides.
We supply a set of dedicated Sisense dashboards, so that you can also create additional dashboards/charts for whatever you need. No scripting or manual API work needed to keep it current. And we're always here to help!
Feel free to reach out in the Community, or here if this is of interest to you, and we'll be happy to get you set up with a demo and a free trial!
Mia from QBeeQ, a Sisense Gold Implementation Partner
Oleksandr_K any news on whether this is coming, in the product?
- Oleksandr_KSisense Employee
dougnewton, thank you for sharing the feedback and submitting this request. This is vital for us!
We will make sure to investigate it and add a straightforward ability to get this information in the UI for the users.Will keep you updated!
Looks like much of this might be available already via the API:
- Get the list of cubes (live and elasticube)
- For each one, get the datasets in that cube
- For each dataset, get the /datasets/id/tables and the metadata about each table
- Output the combination to a CSV file someplace, and make a Usage Analytics "dependency report" on top of it
So just need to wrap all this up in some automated process and expose it in the Sisense product. 🙂
I know I can get some of this from the Usage Analytics (build history), but:
- It doesn't include cubes that aren't built
- It doesn't include live data sources