> For the complete documentation index, see [llms.txt](https://docs.midio.com/midio-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.midio.com/midio-docs/etl-and-high-performance-computing.md).
# ETL and high performance computing
Midio has an **experimental** native package for using the [**polars data frame library**](https://pola.rs/) for handling large amounts of data.
## How to use it
Start by adding the **polars** package using the [package manager](/midio-docs/package-manager.md).
Then, use one of the functions in the `Polars.Source` module, to import your data into the polars format. Supported formats include CSV, JSON, newline delimited JSON and querying a Postgres database.
These functions return a dataframe object, which can be operated on using either `Polars.Execute Sql` or `Polars.Execute Dynamic Sql`.
### Execute Sql
`Execute Sql` lets you provide a list of inputs using, and then executing an SQL query over those inputs. By default its accepts a single input.
### Execute Dynamic SqlAbout the output
Works in a simliar way, but expects an object where the key is the name of the source, and the value is a data frame.
## Getting the results
After executing one or more queries, the results can be collected using the `Polars.Collect` function.
### About the output
The output is by default in a column-oriented format, meaning you get an object where each field represents a column, and its value is a list containing the values for each row.
To get this data converted back to a list of objects, which is often a more useful format, you can use the `Transpose` function.