Tableau – hilllaneconsulting.co.uk

Training

Desktop — downloads

Server — Downloads and top topics

There is a trail version of Tableau Server for both Windows and Unix
https://www.tableau.com/products/server/options

Documentation of Tableau Server Unix Install

https://onlinehelp.tableau.com/current/server-linux/en-us/install.htm

Tech Specs
https://www.tableau.com/products/techspecs#server

Notes

Tableau desktop is used to initially create data sources and connections
Tableau desktop is used to publish content to Tableau Server. Tableau Server can be used to share content with other users.
Tableau server can be installed on Linux despite tutorials stating windows OS only
Seems we are able to incrementally refresh extracts within tableau desktop/server, just need to set the extract refresh type to incremental, and decide on a CDC column.
Tableau server allows AD or local LDAP authentication. Suspect NT will have installed Windows Server product with AD integration ?

Incremental Data Ingest

Incremental Refresh ( desktop ) A simple test using Desktop does prove we can refresh tableau data with incremental datasets from the source system, however looks to only deal with identifying new records based on a match key i.e. primary key. It by default doesn’t seem to deal with performing update/insert operations if an existing key value already exists in both Tableau and the incoming incremental feed.
https://onlinehelp.tableau.com/current/pro/desktop/en-us/extracting_refresh.htm.
Sadly looks as though Incremental refreshes into Tableau only deal with new rows, it cannot merge/update existing rows via a match key, see reference https://www.tableau.com/tableau-data-extracts-part3
However this post suggests that Merge/UpSerts might be possible https://community.tableau.com/docs/DOC-10992

Potential Incremental Ingest Solution

The following process would work if :-

We can identity deltas to extract from SSV via a last update date/last extract date
The dataset contains a primary key on which we can match records to identity changes.

The steps would be :-

Extract the delta dataset from SSV for a “table1”, based on last_update_date / last_extract_date, so we have a CSV/dataset of just the deltas we want to load into Tableau
Extract the current dataset from tableau of “table1”
Union the Delta dataset for “table1” with the Full Extract of the dataset for “table1” within Tableau, where the data in the Tableau dataset does not exist in the Delta Dataset.i.e. SQL could be :-

select * from delta.table1

UNION

select * from tableau.table1

where tableau.table1.ID not in (select ID from delta.table1)

We are now left with a FULL dataset for “table1” which contains the delta from the SSV source system, and the existing Tableau dataset, so we have built the full dataset, combined with the detla. Now refresh the Tableau internal extract with this new dataset.

Can use the API to generate Extract/Refresh