DuckDB

The DuckDB 1.2.2 bugfix release is scheduled to be released in early April.

Since the release of DuckDB v1.0.0 last summer, the time needed to run all TPC-H queries on the 300 GB data set went from 5 minutes to less than 2 minutes! These experiments were run on a MacBook Pro (M1 Max / 64 GB RAM) with the Hyperfine benchmark tool to reduce noise.

Hey folks! We have been collaborating with the team at everyone's favourite in-process SQL OLAP database management system DuckDB to bring you a new podcast series - the DuckDB in Research series! At Disseminate our mission is to bridge the gap between research and industry by exploring research that has a real-world impact. DuckDB embodies this synergy—decades of research underpin its design, and now it’s making waves in the research community as a platform for others to build on and this is what the series will focus on! Join us over the next 5 weeks as we talk with: 📌 Daniël ten Wolde – DuckPGQ, a graph workload extension for DuckDB supporting SQL/PGQ 📌 David Justen – POLAR: Adaptive, non-invasive join order selection 📌 Till Döhmen – DuckDQ: A Python library for data quality checks in ML pipelines 📌 Arjen P. de Vries 🕊️ – FAISS extension for vector similarity search in DuckDB 📌 Haralampos Gavriilidis – SheetReader: Efficient spreadsheet parsing Whether you're a researcher, engineer, or just curious about the intersection of databases and innovation we are sure you will love this series. Subscribe now and stay tuned for our first episode coming Thursday 13th March! 🚀 Trailer Episode: https://lnkd.in/eUYQnnhU

Starting with version 1.2.0, DuckDB supports reading and writing Parquet Bloom filters. This allows DuckDB to efficiently run queries with highly selective filter conditions – directly on the Parquet files. For more details, read the blog post by Hannes Mühleisen https://lnkd.in/dKrTaEWH

If I had a row for every time someone described duckdb as an in-memory database, I would have more rows than would fit into memory.

The recordings of the second DuckDB Amsterdam Meetup are available at https://lnkd.in/dVekk4CD Thanks to our hosts at Xebia and our speakers Chris Laffra, Diederik Greveling, Frank Mbonu and Marco Slot!

New blog post: Gems of DuckDB 1.2 In this post, we explain how we streamlined the installation process of the DuckDB CLI client. We also list several new features, including shorthands (-f), optimizations (IN/OR pushdown), security features (allowed*), and the ability to read/write Excel files using a lightweight extension. Read the full post at https://lnkd.in/d5Skh7aa

DuckDB 1.2.1 is out: https://lnkd.in/dVuPvuua

New blog post by Alex Monahan and Archie Sarre Wood: Reading and Writing Google Sheets in DuckDB This post introduces the gsheets community extension, which enables DuckDB to directly read from and write to Google Sheets. You can replace the convoluted, $ sign-ridden VLOOKUPs with JOINs and make your Google Sheets part of automated pipelines! https://lnkd.in/epkCB9Xd

New DuckDB guide by Gábor Szárnyas: analyze your Git logs using the Fishcake with Swirl emoji 🍥 – https://lnkd.in/dzVZw7sZ (Image attribution: Emojipedia)