How do we get the data from database to Tableau for reporting?
Inorder to get the data from the database we have lot of native connectors you can make use of them.
e.g. To build a one of the chart , I need data to be filtered by date range and couple of other filters. NO I do not see such filters in tableau which can filter the data at database level. I know you can put parameters in custom SQL in Tableau but you can pass only single value and do not see a way to filter when users selects multiple values
check the below link to get an idea how to use custom sql.
IN BO you have universe where all your relationship is defined and it’s a semantic layer but I do not see such concepts in tableau, I guess we either need to define relationship amongst the tables of interest for each dashboard or create custom SQL or views in database to get the data of interest analyses it. It is a correct way and what’s the general best practice - CustomSQL/Database views or Get table relationship defined in tableau.
Answer:In tableau Semantic layer helps to centrally manage the data sources,metadata,calculated fields etc.Data server is managing the semantic layer.So using this semantic layer users can connect to a single data source and work on the same data source for adhoc querying.This shared datasource is something similar to a universe in business object.
1 of 1 people found this helpful
1. In Tableau you create a new connection to the database. This connection could be to raw tables, views, stored procedures, or some combination of the above. With Tableau v10.0 and above you can perform actual joins across different databases, with prior versions you either had to do the integration work before Tableau or use a feature called data blending that acts as a form of post-aggregate left-join (that's a simplification, but a useful one). Check the Tableau documentation for more information.
2. In Tableau you can add fields from the Tableau data source to the Filters shelf. In SQL terms this effectively adds the filters to the WHERE or HAVING clause of the query that Tableau issues, in some cases filters are represented in the SQL as inner joins. Quick Filters include range filters and multi-select. For example here's a view where I've added Order Date and Sub-Category to the Filters Shelf and enabled Quick Filters for them:
2.1. You're right, Tableau does not have a semantic layer in the same way that BO and other tools do. What we can do in Tableau is join a set of tables/views/stored procedures together to create a data source, add additional metadata (comments, assign colors, default formatting, create calculations, etc.) and then publish that data source to Tableau Server so it can be shared amongst multiple workbooks.
In general we're best off using Tableau's native tools to join tables, as this will enable Tableau to most efficiently generate the queries to sources. For example Tableau can do join culling and only query for the necessary tables and fields for a specific view. Custom SQL is really useful for development and necessary in some cases, however in general the recommendation is to avoid it because Tableau is unable to do join culling on Custom SQL (it always executes the entire query). Instead of Custom SQL the alternative is to put the necessary code inside a database view or stored procedure, then you can a) take advantage of the optimizations that the database can do for views & stored procs and b) use the database development tools (Tableau's Custom SQL editor is a very limited text editor.).
Also you might check out a couple of ways of directly connecting Tableau to BO that Tamas Foldi has put together, see http://databoss.starschema.net/accessing-sap-businessobjects-from-tableau-using-web-data-connector/Accessing SAP BusinessObjects from Tableau using Web Data Connector -Databoss for details.http://databoss.starschema.net/accessing-sap-businessobjects-from-tableau-using-web-data-connector/
2.2. Unfortunately the question is too generic to give a good answer here. Are the sources from the same database or different databases, is the merging of sources an inner or outer join or a union, at what point do filters need to be applied, what are the data volumes and the size of the join, etc. are all questions that would need to be addressed before we could give you a recommendation.