Hello, I'm trying out influxdb and would appreciate suggestions on sizing/configuration.
My current load: * metrics are sent using graphite line protocol. * 7 million unique data points per minute, 7 milllion unique measurements (each measurement receives 1 data point per minute) * no read queries for now. When i have configuration that can consume data points successfully query load will be about 16,000/sec. Query example - get all points for the last 24 hrs for a specific measurement. Results that I have so far: * On AWS c4.4xlarge with EBS (gp2, 2.5TB ~ 7,500IOPS), ubuntu 14.04. influxdb uses up all 30GB of RAM and is killed by OOM within first hour. * On AWS c4.8xlarge with EBS (gp2, 2.5TB ~ 7,500IOPS), ubuntu 14.04. influxdb uses about 45GB of RAM and keeps running but it cannot consume all data points at the rate expected by the sender and the sender drops about 2.0mil/min out of 7mil/min. Any recommendations? Thank you, Dmitry -- Remember to include the version number! --- You received this message because you are subscribed to the Google Groups "InfluxData" 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/f533e640-2464-43a5-ace7-1aba8259c5b8%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
