Time series dialects-Lindorm(Lindorm)-阿里云帮助中心

Lindorm SQL provides dialects for LindormTSDB based on the time series data model, including latest value queries, downsampling queries, anomaly detection, continuous queries, and pre-downsampling queries.

Time series dialects

Time series dialects are divided into two main categories: query dialects and precomputation dialects.

Query dialects

Dialect	Description
Latest value query	Queries the latest value in each time series.
Downsampling query	The most common query type for time series data. Aggregates data on the time dimension to reduce the sample rate and decrease query time.
Time series anomaly detection	Detects abnormal data points in a time series to help you promptly identify faults or issues.
Common time series functions	Commonly used functions for querying time series data, including aggregate functions, selection functions, and time processing functions.

Precomputation dialects

Dialect	Description
CREATE CONTINUOUS QUERY	Creates a continuous query task.
DROP CONTINUOUS QUERY	Deletes an existing continuous query task.
SHOW CONTINUOUS QUERIES	Queries existing continuous queries.
CREATE PREDOWNSAMPLE	Creates a pre-downsampling rule.
SHOW PREDOWNSAMPLES	Queries existing pre-downsampling rules.
DROP PREDOWNSAMPLE	Deletes a specified pre-downsampling rule.