on 04-28-2016 4:05 PM
It appears that all of the data sources except "Connect to SAP HANA" will download the entire data set locally. Is this correct?
If I'm not using HANA, how can I navigate large sources without downloading the entire data set locally?
Hi Joe,
That is true, currently Lumira can have a live connection only to SAP HANA. All other data sources (Excel/BW/Universe/SQL) are downloaded.
Things to consider depending upon your datasource,
There is option to save the document without data in case of SAP BW/HANA/UNX.
We can perform a manual refresh/schedule a refresh in the BI launchpad.
The case is slightly different when we load Big Data.
Thank you,
Varun Anand
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
Thank you, Varun.
I was hoping that Lumira would be able to take advantage of universe-based hierarchies to pull down only the data needed to satisfy the current view (similar to how WebI works). It would be difficult to work with very large data sets if it's all processed locally.
I'm going to see about getting a test HANA environment set up so I can experiment with it.
Hi Joe,
When connecting to Universe in Lumira , user has the option to select the required dimensions and measures and also apply filters on then. So this is would be something similar as what we do in Webi wherein we get that data and store in the local cube. This way we should be able to restrict the data to a smaller set in lumira.
Thanks,
Sudarshan
That's not what I was referring to.
Let's say I have a geographical hierarchy in my universe (Region / State / County City), and a bunch of measures (Revenue, Customer Count, etc.). In WebI I create a report with just the Region and Revenue. WebI will generate a query to the source database that just pulls those two columns, so my result set is ~4 rows". Now I drill into the Northeast region. WebI generates a new query to the database, filtered on the Northeast region, and returning one row for each state. And on down to City; so my result set is never larger than the number of cities in a county.
But to achieve the above in Lumira (if I'm understanding things correctly), my initial universe query must include all objects that may be used. So my result set will be significantly larger.
The above is a simple example with a fairly manageable result set, but it's reasonable that I'd want to use Lumira to interactively explore a dataset with billions of rows. It doesn't appear that that is possible unless the data source is HANA.
Hi Joe,
There are 2 versions of SAP Lumira : desktop version and the cloud version.
Desktop version only uses your local computing power while Lumira Hana cloud can handle large volumes of data efficiently.
Thanks and Regards,
Deepa.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
User | Count |
---|---|
85 | |
10 | |
10 | |
10 | |
7 | |
6 | |
6 | |
5 | |
4 | |
4 |
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.