Scientific Computing in Rust Monthly #14

Welcome to this month’s issue of Scientific Computing in Rust Monthly and happy 2026 to everyone!

Scientific Computing in Rust Monthly can be viewed online at scientificcomputing.rs/monthly, you can sign up to receive an email copy at mailinglist.scientificcomputing.rs, or you can subscribe using the RSS feed at scientificcomputing.rs/monthly/rss.xml.

Items can be added to the next issue of the newsletter by opening a pull request to the Scientific Computing in Rust Monthly repository on GitHub. You can find more information about adding items here.

Crate of the month: measurements

crates.io ~ docs.rs ~ source

The measurements crate allows you to represent physical quantities. You can get your quantities in different, compatible units and perform arithmetic operations. measurements is no-std compatible. Thus, you can even use proper units in your embedded projects! This crate has a small footprint and is highly useful to avoid common errors of, e.g., forgetting unit conversions, which can be very costly (e.g., Mars Climate Orbiter).

New releases

Bevy 0.18.0

Bevy is a data-driven game engine built in Rust. This release includes numerous new features, bug fixes, and quality-of-life improvements. A detailed blog post can be found here.

Burn 0.20.0

Burn is a tensor library and deep learning framework optimized for numerical computing,m model interference, and model training. This release features major performance improvements thanks to the introduction of CubeK.

InterpN 0.11.0

InterpN is an n-dimensional interpolation/extrapolation methods library that is no-std compatible. This release provides some behind-the-scenes improvements. Notably, all methods are now compatible with static analysis of memory usage and early bounds checks were improved.

rlst 0.6.1

The Rust Linear Solver Toolbox rlst provides dense and sparse linear algebra routines in Rust. Since the last newsletter, version 0.5 and 0.6 have been released, adding various methods, functions, and interfaces. Details can be found in the release notes here.

Publications

Irving B., et al., Mantaray: A Rust Package for Ray Tracing Ocean Surface Gravity Waves, Journal of Open Source Software, 10(116)

Mantaray is a open-source Rust package with a Python interface that solves the ray equations for ocean surface gravity waves. It is designed for performance, robustness, and ease of use. Mantaray can currently be applied to idealized and realistic wave propagation problems.

Wenger, A. (@fasterthanlime), The science of loudness

In this blog post, Amos walks the reader through the physics of the science of sound, loudness, noise, and the associated units. The follow up article Making our own spectrogram builds on this and shows how to make a spectrogram in Rust. This is a very accessible introduction and contains many fun animations and code snippets.

Help wanted

Goombay-rs

Goombay-rs is the Rust implementation of my Goombay library. Goombay is a sequence alignment library with over 20 algorithms written entirely in python. My initial goal with Goombay-rs is to implement the algorithms that are present in the python version of the library. The secondary goal of the library would be to speed up computation time where possible while still maintaining the core fundamentals of each algorithm. A tertiary goal of the project is to include the ability to pip install the project to be used in Python. Spindalis (which also has open issues) is the primary dependency for Goombay-rs used for matrix manipulation.

Miscellaneous

CocoIndex

An open-source, data transformation framework with its core engine written in Rust, designed for building and maintaining continuously updated indexes over large datasets. CocoIndex supports incremental processing and provides data lineage tracking, making it well-suited for scientific workflows involving evolving research data. Potential applications include indexing academic papers with structured metadata extraction, building knowledge graphs from research outputs, and creating semantic search capabilities over scientific literature. The framework uses a dataflow programming model and integrates with various databases including PostgreSQL and vector databases for semantic search applications. Community discussion and support available on Discord.