Ursa Labs Blog · Ursa Labs

Conbench is a language independent, continuous benchmarking framework, built specifically with the needs of a cross-language, platform-independent, high-performance project like Arrow in mind.
Because speed and efficiency are important design goals of Apache Arrow, we've been working on tools to help us investigate and monitor performance. Here, we introduce our R package for benchmarking and use it to revisit some previously reported CSV benchmark results.
Announcing Ursa Computing, a startup to empower data teams and accelerate data science, machine learning, and AI work.
In this blog post we announce the grant we have received from the Chan Zuckerberg Initiative to create an open source apprenticeship program for Apache Arrow with the objective of promoting diversity and inclusion among the maintainers.
In this blog post we introduce the new Feather V2 file format and revisit file format performance benchmarks from last October
In this blog post, we discuss the work we've done in Apache Arrow to efficiently construct pandas DataFrame objects. In some cases, we can do it faster than pandas itself.
In this blog post, we take a look back at our work in 2019 and give insight into our development roadmap for 2020 and beyond.
We are excited to announce that Ursa Labs has several new sponsors supporting our work on Apache Arrow over the coming year:
With a busy fall full of development and some travel I wasn’t able to keep up with monthly or bi-monthly reports, so this report gives highlights about the Ursa Labs team’s work from August up until now. With 2019 nearly behind us, this also gives us a moment to reflect on everything that’s been accomplished this year and look forward to next year. We hope to write a blog post soon discussing the big picture plan ahead for 2020.
In this blog post I do some performance investigations of a few binary columnar formats (Parquet, Feather, and FST) in Python and R