It’s one of the big buzz words of SAP. The database/application is of course known for its speed due to its in memory technology. Microstrategy is positioning itself as an ideal tool for unlocking reporting and mobile with HANA certification and all.
Having done dashboarding from Microstrategy on SAP BW, I was curious to see how this would work out.
Here we are using SQL to connect directly to the HANA database. This is a big difference from connecting SAP BW running on HANA where the less preferable method of MDX is still needed.
When trying to get the data into Microstrategy I encountered some difficulties.
Mapping and previewing the data worked fine but when executing any report it returned an error that the table was not found. This lead me the the technote TN: 40920 but no fix was mentioned.
As a workar0und, I tried to create a SQL statement in HANA studio directly on a calculation view, which I then copy as a freeform SQL into Microstrategy.
After playing around for a bit I found that selecting data preview and subsequently log, shows the SQL statement used for the calculation view generation.
Copying the SQL into a Microstrategy freeform report runs correctly. Of course a big downside is the added effort and upkeep of using freeform.
I decided to go over the webcast again from Adobe running Microstrategy HANA. Here it is mentioned to use import table prefix when using the warehouse catalog editor, which I did not do initially.
After testing an attribute on the new table, the report ran without any problems.
Tip: It shows columns as tables as well in my case. Trying to report on these results in errors. The table name should give a clear indication if this is an actual table.
Having now created the connection you can build a schema like on any other database.
Although HANA is of course known for fast performance, in this case the results were disappointing, with a run time of 19 seconds for retrieving 5000 records, using HANA Studio on my Laptop. I expect this might have to do with the network as we are using a HANA cloudshare server and reports on a remote desktop session return significantly faster.
Running a Micostrategy report on the Calculation view seems slightly slower by 1-2 seconds then freeform SQL, this might be due to table size, or the fact it is actually running another SQL statement in the background.
To summarize In short I was pleasantly surprised by how ease of building reports on top of HANA, also using different HANA views. This is definitely an improvement over using MDX.
