3 ms·
Thanks for making this! I think making them columns instead of putting them in the jsob column would make sense. Like we have a few top level properties that w
by bni 9y ago
Thanks for making this!
I think making them columns instead of putting them in the jsob column would make sense. Like we have a few top level properties that we currently use scan (ugh!) to generate some reports.
This is clearly a bad design and not what dynamodb should be used for. So using Podyn we could definitely improve this and having the top level properties as columns I think would make migration easier for a lot of people.
I have actually considered writing my own code (AWS Lambda) to do a sync like this, now I guess I won't have to.
Would it be feasible to make Podyn run in a scheduled Lambda function? would save us from having an EC2 around...
- mslot 9y ago> Would it be feasible to make Podyn run in a scheduled Lambda function? would save us from having an EC2 around... That's an interesting idea! My mental model for running Podyn was a container or T2 instance, but I think it wouldn't be hard to add Lambda support. I've created an issue for tracking it: https://github.com/citusdata/podyn/issues/2 https://github.com/citusdata/podyn/issues/2