News for hackers. Stuff that matters.

A comment on Ask HN: How to store billions of rows of time-series data?.

Re:Ask HN: How to store billions of rows of time-series data? by valyala on Monday April 06, 2020 @04:23PM (#22797428)
Take a look at VictoriaMetrics. It supports OHLC function (rollup_candlestick) via MetticsQL [1], which is backwards compatible with PromQL - query language used in Prometheus. PromQL is optimized for typical queries over time series data [2].

VictoriaMetrics works out of the box without the need to set up third-party systems or wroting complex configs. It is also very fast for both data ingestion and querying [3], [4].

[1] https://github.com/VictoriaMetrics/VictoriaMetrics/wiki/Metr...

[2] https://medium.com/@valyala/promql-tutorial-for-beginners-9a...

[3] https://medium.com/@valyala/measuring-vertical-scalability-f...

[4] https://medium.com/@valyala/billy-how-victoriametrics-deals-...

[ Parent | on HN ]

There are no comments on this comment yet.