Data Fusion Capabilities

Has anyone used Data Fusions extensively? I've created a data fusion that combines 4 data sets and in total has about 14 million rows of data at this point across about 120 columns. The reason I choose data fusion was to allow for my team to reprocess data if needed should we find bugs in the data we ingest, to backdate easily if we wish to add new columns to the raw data, and to allow changes made in our salesforce instance to update the data tagging instantly rather than having to rerun the entire data set. However I'm finding some issues with Data Fusion around performance of the cards I build on it. Drill downs don't load very well or time out constantly which renders the cards I build limited in impact. 

 

Does anyone else have these issues with Data Fusion? Also are you an enterprise client or not since I believe there is a difference in processing power provided to enterprise clients.

Comments

  • Have you tried creating a dataflow off your resultant dataset?  It may help the performance of your cards if you do some light etl on the result to lighten the load. 

     

    You may need it to hit all 14M rows, but likely aren't using 120 columns in your cards. 

     

    If you add a column in the future that you need accessible in your cards, it will only take a few seconds to add that column to your output, and it'll save you loads of time and frustration from your cards not loading well.