It look very interesting, and I am a bit surprised to see how a tree structure can deliver such high performances in retrieving the data. Querying a long period of time mean that each top node has to be walked to the bottom, right? Isn't that fairly expensive?
The aggregations on node level are also interesting. It is something that also the company I work for (https://www.energy21.com) does for calculations, but we use a relational database and a notification-based system for performing the recalculations and keeping the aggregations up to date.
https://blog.acolyer.org/2016/05/04/btrdb-optimizing-storage...
https://pingthings.io/platform.html
It look very interesting, and I am a bit surprised to see how a tree structure can deliver such high performances in retrieving the data. Querying a long period of time mean that each top node has to be walked to the bottom, right? Isn't that fairly expensive?
The aggregations on node level are also interesting. It is something that also the company I work for (https://www.energy21.com) does for calculations, but we use a relational database and a notification-based system for performing the recalculations and keeping the aggregations up to date.
Cool stuff.