When comparing KairosDB vs VictoriaMetrics, the Slant community recommends VictoriaMetrics for most people. In the question“What are the best time-series databases and/or data stores?” VictoriaMetrics is ranked 4th while KairosDB is ranked 6th. The most important reason people chose VictoriaMetrics is:
VictoriaMetrics requires fewer hardware resources - RAM, CPU, storage - compared to competitors. This allows for saving hardware costs.
Ranked in these QuestionsQuestion Ranking
Pros
Pro Apache Cassandra provides fast, efficient, and highly reliable storage
KairosDB primary storage backend is Apache Cassandra. Cassandra is one of the biggest player in NoSQL world, supported by Datastax and used by large companies such as Netflix.
Pro Open source
KairoDB is licensed under Apache License with source code available on GitHub.
Pro High modularity
Extending KairosDB is very easy.
Software is based on Guice modules and all parts can be extended using modules: aggregations, query engines, webservice APIs.
Pro Multi apis
Accessible through Rest, Telnet, and Graphite.
Pro Fast and cost-effective
VictoriaMetrics requires fewer hardware resources - RAM, CPU, storage - compared to competitors. This allows for saving hardware costs.
Pro Easy to setup and operate
Just start a single binary with the minimum amount of settings.
Pro Different protocols support
Supports metrics' scraping, ingestion and backfilling via the different protocols - see here.
Pro Seamless integration with Prometheus+Grafana
Prometheus may be configured to write data to VictoriaMetrics. Grafana may be configured to query the data directly from VictoriaMetrics using PromQL.
Pro Open source
VictoriaMetrics is free and open source - see here.
Pro Cluster version is free and open source
Cons
Con Built from scratch
It is young and it is built from scratch. This means it can contain unpolished code.
Con Doesn't support SQL
It supports only PromQL for querying data.