Looking to use MS Power BI to generate dashboards. Anyone familiar with connecting to an E11 cloud instance with Power BI? I am trying to use a product from Cdata which does connect however it does not provide data. I am working with their support team to try to resolve but would like to know if there are other options.
Thanks…
What do you mean it does not provide data?
I experimented with CData about six months ago. It often timed out and was extremely slow when it worked. And IIRC, it requires you to set up a separate virtual server for every REST service (SalesOrder, etc.) so you cannot simply join OrderHed to Customer or something like that. It’s nowhere near a viable alternative to direct SQL access.
1 Like
When using the cloud, especially BI products, one should consider what data one wants to analyze. One has to balance the cost of computation against the cost of transit. In an on-prem world, we are fantastically inefficient and query the database constantly because the cost of transit is cheap. Hell, the cost of compute is cheap too. If you own the power station, you’ll leave lights on all night and day. But in the cloud (where you are paying for cycles), one has to be more efficient.
Are we talking about dynamic, up-to-the-minute data like: shipments today, who’s logged into a job, what jobs are open right now, etc. If so, BAQs with EDD or a custom solution in ASP.NET 6 is tough to beat.
Are we talking about immutable data like G/L postings? Since PartTran, GLTranDtl, posted AR/AP groups cannot be changed once posted, they become great candidates to transfer to place closer to the BI engine to get better performance and not waste all that time querying the original source again, and again, and again… Point in time data fits this as well. For example, who’s logged in right now and how many licenses are they using. That’s immutable and can be posted to a data store that’s more easily analyzed by the BI tool.
So think about your use cases and choose appropriately. One tool does not rule them all, IMHO.
4 Likes
Love it Mark!