0

I'm beginning the process developing a dataset that will help me track changes made to records in our MySQL backend. More specifically, I want to be able to report on the original record value, the new record value and the timestamp for when the change occurred. Here is an example:

  • Record X = 20.
  • A change is made to Record X on March 26th so it no longer = 20.
  • As of March 26th, Record X = 30.

In this scenario, the original record value (for Record X) was 20, the new record value = 30 and the change occurred March 26th, 2014.

I considered taking daily snapshots and storing the snapshot data in GoodData's warehouse. This would allow me to make daily comparisons of the record values. The problem with this approach is that our consumption of warehouse storage space will increase dramatically and at a very fast rate. I would prefer to track this data only when changes occur (incrementally). I don't know exactly how to get started here. Can someone point me in the right direction?

phatskat
  • 1,797
  • 1
  • 15
  • 32
user3464868
  • 21
  • 3

1 Answers1

0

from what I see the best approach for you if you are going to use CloudConnect is to create a query to your MySQL database (use DB Reader component) to select data that has been changed from the last upload (from yesterday). Then you can use the incremental loading to get data into the GoodData. And have some kind of snapshot-id or snapshot-date connected.

Can you share some more details about your use case? What exactly you would like to track finally? What is being shown in dashboard?

Thanks for asking! Regards,

JT

Jiri Tobolka
  • 635
  • 3
  • 13