News for hackers. Stuff that matters.
Re:InfluxDB vs. Cassandra for timeseries data by dz0ny on Thursday September 08, 2016 @06:03AM (#12451634)
It would be interesting to compare memory requirements, I chose Influxdb because it had 10 times lower memory usage. The dataset was small (couple of million datapoints)... but stil
Re:InfluxDB vs. Cassandra for timeseries data by dx034 on Thursday September 08, 2016 @06:44AM (#12451781)
That only works when you have one series with a lot of observations. If you have many series with fewer observations (say 50k per series) influxDB uses absurd amounts of memory. I had to switch back to Cassandra because I constantly ran out of memory.
Re:InfluxDB vs. Cassandra for timeseries data by pauldix on Thursday September 08, 2016 @06:52AM (#12451816)
We're working on solving the high cardinality problem. Hopefully soon
Re:InfluxDB vs. Cassandra for timeseries data by deluvas on Thursday September 08, 2016 @08:20AM (#12452279)
How much memory are we talking about? How often did you execute queries on the data?

I'm asking because my first impression of Influxdb involved lots of memory gobbling.