Get better performance for complex retrieves

0
Hi, Our app has a functionality for exporting reports in excel format. Number of rows is not that big (just  under 2k at the moment) and 125 columns. For each row/client there are multiple and complex retrieves and calculations that build up to over 5 minutes. I introduced batches to the flow. It turned out that batches of 100 have the best performance in our case, but in general it didn't have a huge impact. Now it takes 1 minute less than before.  Maybe some of the great minds here can suggest any ideas about how I could improve performance? Apart from the time, this functionality also takes up a lot of memory and will crash in some exceptional circumstances when the load is larger. Simple caching or saving of the reports won't work, because data can be changed and this functionality was particularly built for retrieving the most accurate data. Anything else?  Also it might be relevant to mention that we will be upgrading from version 8 to 9 soon. Do you think it might change anything in my case?
asked
1 answers
1

Iryna,

This academy content might be a good place to start in optimizing the XPath you are using:  https://academy.mendix.com/link/modules/22/lectures/180/4.3-XPath-Optimizations

 

In addition, if you are building up this content via a microflow, you could have a look at the suggestions on this page:  https://docs.mendix.com/refguide/community-best-practices-for-app-performance/

 

Hope these help you improve your performance!

 

Mike

answered