RRDTool VS Cortex Project

For anyone interested in how the graphs were made: https://oss.oetiker.ch/rrdtool/ Happy to see people still using RRD after all these years. - Source: Hacker News / over 1 year ago

(I actually implemented the firmware for this... Using the Arduino port for ESP8266 because I was impatient, but discovered that either my DHT11 modules are junk or I'm misusing them, because the humidity measurement drifts as they keep running. I need to write the Rust+rrdtool app meant to receive the reports and then do some comparative tests between the DHT11 and some BME280s, and between the current firmware... Source: about 2 years ago

Nice! This is giving me some ideas. Here's what my old school rrdtool-based system looks like:. Source: over 2 years ago

Now if its more metric data you are using and want to do APM, prometheus is your man https://prometheus.io/, want to make prometheus your full time job? Deploy cortex https://cortexmetrics.io/, honorable mention in the metrics space, Zabbix, https://www.zabbix.com/ I've seen use cases of zabbix going way beyond its intended use its a fantastic tool. Source: 11 months ago

Yes, but also no. The Prometheus ecosystem already has two FOSS time-series databases that are complementary to Prometheus itself. Thanos and Mimir. Not to mention M3db, developed at Uber, and Cortex, then ancestor of Mimir. There's a bunch of others I won't mention as it would take too long. Source: 11 months ago

You can use the Remote write feature to send to a centralized location. It would have to be scalable like Cortex https://cortexmetrics.io/. Source: about 1 year ago

For a homelab I think prometheus + grafana is easy to get started and scales well. There are lots of ways to set up the architecture. Prometheus can write to a directory on a filesystem, it can be set to write to a remote server, and there are other projects to integrate object storage (s3, minio, etc) or influxdb for long term storage and downsampling. Source: over 1 year ago