Catalogue
/
Databases
/
DuckDB for Lightweight High-Performance Analytical SQL

DuckDB for Lightweight High-Performance Analytical SQL

A practical two day technical course for participants who are new to DuckDB and want to use high-performance analytical SQL directly within local, data engineering, and Python workflows. The agenda progresses from DuckDB architecture and analytical SQL through direct file querying, Parquet integration, Python interoperability, and performance optimisation. The course focuses on using DuckDB effectively as an embedded analytical engine without introducing unnecessary infrastructure.

What will you learn?

You will learn how DuckDB executes analytical workloads, how to query local and remote datasets directly with SQL, and how to work efficiently with Parquet and other columnar data. You will integrate DuckDB with Python and common data processing libraries and understand when to use persistent databases, in-memory processing, or direct file access. You will also learn how to analyse query execution and tune workloads for efficient use of CPU, memory, storage, and external data.

  • Use DuckDB and analytical SQL for complex data exploration and transformation
  • Query, transform, and export Parquet datasets efficiently
  • Integrate DuckDB with Python and common DataFrame and Arrow workflows
  • Analyse and optimise DuckDB workloads for performance and resource efficiency

Requirements:

  • Technical background in software development, data engineering, analytics, data science, or a related field
  • Basic SQL knowledge
  • Basic Python knowledge is helpful for the integration sections
  • Previous DuckDB experience is not required

Course Outline*:

*We customize the course outline and content to your specific needs and relevant use cases.

Module 1: DuckDB architecture and analytical processing

  • Understanding DuckDB as an embedded analytical database engine
  • Comparing in-process analytics with client-server database architectures
  • Working with in-memory and persistent DuckDB databases
  • Understanding columnar execution, vectorised processing, parallelism, and common DuckDB use cases

Module 2: Analytical SQL with DuckDB

  • Creating tables, views, and reusable analytical queries
  • Working with aggregations, joins, common table expressions, and window functions
  • Using lists, structs, maps, and other data types for analytical workloads
  • Applying DuckDB SQL features to data exploration and transformation

Module 3: Querying files directly

  • Querying CSV, JSON, and Parquet files without loading them into database tables first
  • Working with multiple files, file patterns, and partitioned datasets
  • Creating tables and views from external files when persistence is useful
  • Deciding between direct file access and loading data into DuckDB storage

Module 4: Parquet integration and analytical data workflows

  • Reading and writing Parquet directly from DuckDB
  • Understanding columnar storage, row groups, compression, and Parquet metadata
  • Using filter and projection pushdown to reduce unnecessary data reads
  • Querying partitioned and remote Parquet datasets efficiently

Module 5: DuckDB with Python

  • Connecting to in-memory and persistent DuckDB databases from Python
  • Executing parameterised SQL and retrieving query results
  • Querying Python objects and DataFrames directly from DuckDB
  • Managing connections and structuring reusable analytical workflows

Module 6: DataFrame and Arrow integration

  • Moving data between DuckDB, pandas, Polars, and Apache Arrow
  • Querying DataFrames directly with SQL without unnecessary data copies
  • Choosing between SQL and DataFrame operations for different transformation tasks
  • Combining Python application logic with DuckDB analytical processing

Module 7: Query analysis and performance optimisation

  • Using EXPLAIN and EXPLAIN ANALYZE to understand query execution
  • Reducing scanned data through filtering, projection, sorting, and appropriate Parquet organisation
  • Configuring threads, memory limits, and spill-to-disk behaviour for different workloads
  • Recognising inefficient joins, excessive memory use, unnecessary data movement, and other performance problems

Module 8: Designing efficient DuckDB analytical workflows

  • Selecting between persistent tables, in-memory data, and direct file queries
  • Designing repeatable transformations from raw files to analytical outputs
  • Working with larger-than-memory datasets and remote data sources
  • Building a practical checklist covering data layout, connections, performance, reproducibility, and maintainability

Hands-on learning with expert instructors at your location for organizations.

3,072€*
Graph Icon - Education X Webflow Template
Level:
intermediate
Clock Icon - Education X Webflow Template
Duration:
14
Hours (days:
2
)
Camera Icon - Education X Webflow Template
Training customized to your needs
Star Icon - Education X Webflow Template
Immersive hands-on experience in a dedicated setting
*Price can range depending on number of participants, change of outline, location etc.

Master new skills guided by experienced instructors from anywhere.

2,037€*
Graph Icon - Education X Webflow Template
Level:
intermediate
Clock Icon - Education X Webflow Template
Duration:
14
Hours (days:
2
)
Camera Icon - Education X Webflow Template
Training customized to your needs
Star Icon - Education X Webflow Template
Reduced training costs
*Price can range depending on number of participants, change of outline, location etc.

Upcoming Sessions

7-8 Jan 2027
Zurich
28-29 Apr 2027
Stockholm
5-6 May 2027
Brussels
2-3 Jun 2027
Barcelona
15-16 Jul 2027
Amsterdam

Can't find a suitable date? Get in touch and we'll arrange one that works for you.

We use cookies to improve site navigation, analyse how the site is used, and support our marketing. You can accept all cookies, reject non-essential ones, or choose individual categories. Read our Cookie Notice