Update: By reducing the interval between CQ runs in the influxdb config I was able to reduce the constant CPU load from a stable 40-60% to a managable 4-9% when not having queries from external systems come in. For now this works, but in the future I wish to have a database structure that can scale better.
Because I haven't had any replies to my original post I've been bust setting up one of the proposed database structures on a different server instance. Interesting to note is that both the old and the new server are being fed the same data. The new database has four measurements (temperature, humidity, co2, movement) and uses the tag 'd' as the device ID. Instead of six continuous queries per measurement, I now have 12 continuous queries in total! In theory my CPU load should be lower than in the old situation, right? Nope, average CPU load on both servers is about the same (with the old server also servicing clients that query at random as well!). Now I must note that the 'new server' config has not been modified to run CQ's every 5s instead of every 1s, I'm going to try that today. I'm still very interested in whether anyone with experience with higher volumes of data on InfluxDB or anyone on the InfluxData team can tell me whether the CPU load will double if I double the number of data-sources (= tag values) that would be awesome! -- Remember to include the InfluxDB version number with all issue reports --- You received this message because you are subscribed to the Google Groups "InfluxDB" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at https://groups.google.com/group/influxdb. To view this discussion on the web visit https://groups.google.com/d/msgid/influxdb/71b998ab-2078-4c46-a0ce-157f57565190%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
