WIP sql query guide
parent
d31105ab9c
commit
72cc88e814
|
|
@ -0,0 +1,62 @@
|
|||
---
|
||||
title: Use SQL data
|
||||
seotitle: Use SQL data with InfluxDB
|
||||
description: >
|
||||
placeholder
|
||||
v2.0/tags: [query, flux, sql]
|
||||
menu:
|
||||
v2_0:
|
||||
parent: How-to guides
|
||||
weight: 210
|
||||
---
|
||||
|
||||
|
||||
- Import Flux's `sql` package and use `sql.from` to query data from a SQL database such as Postgres or MySQL.
|
||||
|
||||
### Benefits
|
||||
- Some data schemas are better suited for SQL databases. With the ability to query a SQL data source,
|
||||
you can store SQL-suited data there and join it with your time series data.
|
||||
This allows you to reduce your InfluxDB schema complexity and improve performance.
|
||||
-
|
||||
|
||||
## Use cases
|
||||
|
||||
### Reduce cardinality by storing tag data in a SQL data source
|
||||
High series cardinality can lead to high memory usage, higher hardware costs, and impact the overall performance of InfluxDB.
|
||||
The primary culprit behind cardinality is unique tag values.
|
||||
Using a SQL data source, you can offload much of your tag data to the SQL database.
|
||||
As long as you have at least one common tag on which to join in InfluxDB, that tag data can still be associated data in InfluxDB.
|
||||
|
||||
For example:
|
||||
|
||||
#### Join relation data with time series data
|
||||
- Sensor data with relational sensor information
|
||||
- SensorID
|
||||
- Name
|
||||
- Type
|
||||
- Location
|
||||
- Model
|
||||
- Sensor metrics stored in InfluxDB, each with a `sensorID` tag.
|
||||
Each type of sensor metric is stored in a different measurement.
|
||||
- air_quality
|
||||
- temperature
|
||||
- humidity
|
||||
- co2
|
||||
- methane
|
||||
- light
|
||||
- uv
|
||||
|
||||
|
||||
```js
|
||||
import "sql"
|
||||
|
||||
sensorInfo = sql.from(
|
||||
driver: "postgres",
|
||||
driverName: "",
|
||||
query: ""
|
||||
)
|
||||
|
||||
sensorMetrics = from(bucket: "sensors" )
|
||||
|> range(start: -1d)
|
||||
|
||||
```
|
||||
Loading…
Reference in New Issue