1. 4

  2. 4

    Their argument seems to be “we’ve done this numerous times, and therefore we do it better than you”, which presumes that there’s no innovation left here by anyone else. I don’t buy this.

    1. 2

      Same. I’m not even convinced they’ve done this once.

      A time-series database isn’t a graph of hard-drive usage or network rates: “You Can Lose a Few Datapoints Here and There” isn’t true for a lot of applications, so as soon as you’ve moved into the space where you don’t care about keeping correct data, talking about “queries-per-second” or “inserts-per-second” just makes you look like a fool.

      Disclaimer: I actually have built a time series database (twice).