Used Redshift for our large data needs. No reason to look to anything else, given our time-shifted data could easily be fit into Redshift. For larger data, we would have to have some other solution.
This is extremely interesting and I'm eager to see the follow-up.
I'm currently planning something very similar at my company and any reference to how other smaller-scale companies handle this when scaling is invaluable. Are there any other blogs from other companies which have talked about similar things?
These need to be sorted by publishing time, not by your database entry time. You can tell because you've recently added Percona, and it's filling up the top with backdated stuff. I'd also have preferred just a minimalist interface. We don't need anymore voting/commenting stuff at this point.
Interesting feedback. Not sure how I could make the interface more minimal, it's pretty stark as it is.
The backdated stuff is weird I'll admit, but they also posted a bunch of things in the last couple days so could be screwing with the ranking algo. I'll check it out.
> The tool we used did not support using Hive as a backend, and given the speed of even relatively basic Hive queries, it wouldn’t make much sense to use that for reporting metrics.
Did Reddit try setting up Presto? Analytics, not just visualization, with Hive is really hard.
