Hi,

@Joe: you're right, it would be tedious to store the state, and probably
isn't the best route. The ingestion rate of data is one point per 0.1 sec
per device
@Thad: I really like your suggestion, it makes my life a little bit easier.
I'm going to use MongoDB and perform upserts thanks to the PutMongo
processor that already exists. It really streamlines my flow

Thanks both for the suggestions!

Cheers,
Stephane

On Fri, May 27, 2016 at 12:59 AM Thad Guidry <[email protected]> wrote:

> BTW, the idea previously is a Batching Processor, similar to what Spring
> Batch and other Data Tools provide out of the box.  Not sure if NiFI
> already has that concept in one of the processors , or if you have to
> resort to Groovy or the ExecuteScript processor.
>
> Thad
> +ThadGuidry <https://www.google.com/+ThadGuidry>
>
>

Reply via email to