Effective query execution

Scaling and Optimizing Data Pipelines with Polars

Liam Brannigan

Data Scientist & Polars Contributor

Meet your instructor

Liam Brannigan, Lead Data Scientist
ML and Data Engineering Specialist
Polars contributor

Picture of Liam Brannigan - the instructor of this course

From laptop to cloud

Polars pipeline scaling from laptop to cloud

Is this course for you?

Polars course pages

Creating a lazy query with scan_csv
Writing expressions with filter, select, and group_by

Chapter 1 - optimization

Query optimization

Chapter 1 - optimization

Query optimization

Chapter 2 - effective I/O

Image of Polars ingesting multiple file types

Chapter 3 - richer dtypes

Diagram of nested data transformed into a table.

Chapter 4 - scaling pipelines

Image of large table being streamed.

Chicago requests dataset

The Mayor of Chicago's data analytics team

Inspecting the dataset

requests = pl.scan_csv("311_Service_Requests.csv",try_parse_dates=True)

A massive dataset, recording every service request from Chicago's citizens

Inspecting the dataset

requests.collect()

Inspecting the dataset

requests.collect().head(5)

shape: (5, 39)
| TYPE                          | STATUS    | DEPARTMENT     | CREATED_DATE        | ... |
| ---                           | ---       | ---            | ---                 | --- |
| str                           | str       | str            | str                 | ... |
|-------------------------------|-----------|----------------|---------------------|-----|
| Pothole in Street Complaint   | Completed | Transportation | 2019-12-16T10:09:08 | ... |
| Tree Trim Request             | Cancelled | Sanitation     | 2019-09-18T01:05:08 | ... |
| Garbage Cart Maintenance      | Completed | Sanitation     | 2021-01-24T09:14:58 | ... |
| Pothole in Street Complaint   | Completed | Transportation | 2019-03-21T10:41:01 | ... |
| Recycling Pick Up             | Completed | Sanitation     | 2021-02-16T08:28:59 | ... |

Limiting rows processed

requests.head(5)

Limiting rows processed

requests.head(5).collect()

shape: (5, 39)
| TYPE                            | STATUS    | DEPARTMENT     | CREATED_DATE        | ... |
| ---                             | ---       | ---            | ---                 | --- |
| str                             | str       | str            | str                 | ... |
|---------------------------------|-----------|----------------|---------------------|-----|
| Pothole in Street Complaint     | Completed | Transportation | 2019-12-16T10:09:08 | ... |
| Tree Trim Request | Completed | Sanitation     | 2019-09-18T01:05:08 | ... |
| Garbage Cart Maintenance        | Completed | Sanitation     | 2021-01-24T09:14:58 | ... |
| Pothole in Street Complaint     | Completed | Transportation | 2019-03-21T10:41:01 | ... |
| Recycling Pick Up               | Completed | Sanitation     | 2021-02-16T08:28:59 | ... |

Delay calling collect

Department summary query

completed_by_department

Department summary query

completed_by_department = requests

Department summary query

completed_by_department = requests.filter(
    pl.col("STATUS") == "Completed"
)

Department summary query

completed_by_department = requests.filter(
    pl.col("STATUS") == "Completed"
).collect()

Department summary query

completed_by_department = requests.filter(
    pl.col("STATUS") == "Completed"
).collect().group_by("DEPARTMENT").len()

Department summary query

completed_by_department = requests.filter(
    pl.col("STATUS") == "Completed"
).group_by("DEPARTMENT").len().collect()

shape: (10, 2)
| DEPARTMENT                     | len     |
| ---                            | ---     |
| str                            | u32     |
|-------------------------------|---------|
| 311 City Services              | 4859161 |
| Sanitation                     | 3406631 |
| Aviation                       | 2337842 |
| CDOT - Department of Transport | 1468240 |

Diverging query branches

completed_by_department = requests.filter(
    pl.col("STATUS") == "Completed"
).group_by("DEPARTMENT").len()

completed_by_month = requests.filter(
    pl.col("STATUS") == "Completed"
).group_by("MONTH").len()

How the team runs this today

completed_by_department.collect()
completed_by_month.collect()

Executing diverging queries

results = pl.collect_all(


)

Executing diverging queries

results = pl.collect_all([
    completed_by_department,
    completed_by_month,
])

Executing diverging queries

results[0]  # completed_by_department
shape: (10, 2)
| DEPARTMENT           | len     |
| ---                  | ---     |
| str                  | u32     |
|----------------------|---------|
| 311 City Services    | 4859161 |
| Sanitation           | 3406631 |
| Aviation             | 2337842 |
| Transport            | 1468240 |

results[1]  # completed_by_month
shape: (12, 2)
| MONTH | len  |
| ---   | ---  |
| i64   | u32  |
|-------|------|
| 1     | 2506 |
| 2     | 4566 |
| 3     | 2739 |
| 4     | 2922 |

Let's practice!

Scaling and Optimizing Data Pipelines with Polars