Question, Unstructured events and CSV import

For data modelers & consumers Redshift
1

Does it has sense to send unstructured events to import data thru a CSV ? the reason is because I already have previous data that needs to be loaded in there as a “Migration” and later start using the tracker in order to add new data.

And to import data I was about to create a CSV file and use nodejs-tracker to send unstructured events to the collector, but I am afraid because the file contains more than 4 millions of rows.

If there is some trick to add into S3 and load all the events at once using just the etl runner, that would be awesome.

2

Hi @Germanaz0 - this sounds like a good use case for a combination of:

It would be super-awesome if you would blog your experience, or share it in this thread as a tutorial!

3

Sure I will, going to check how it goes, with node-js works fine but I think that the CLI would work much faster

4

The nice thing about the CLI is you can script up a bash file that does increasingly large batches of your events, as you gain confidence that the events are making it through to your collector…

5

Yes the only con that I’ve found is that there is no buffering, in some point is safer without a buffer, but slow response