Anubis and Marauder’s Map: A Deep Dive Into Your Code Database

The idea behind this tool is to represent the evolution of your codebase targeting specifically High-Performance Computing (HPC) softwares. It uses codemetrics to analyze complexity, technical debt, and workforce dynamics at work in scientific computing codes. It highlights the challenges posed by legacy systems, such as keeping affordable maintenance costs or adapting to modern GPU-powered supercomputers, and introduces tailored tools like Anubis to evaluate codebase structure and developer interaction.             

Description

Codemetrics involve quantitative methods to assess software quality, focusing on factors like code complexity, maintainability and developer collaboration. These metrics help identify technical debt, inefficiencies, and areas for improvement, guiding better software design and management practices. By applying codemetrics, teams can evaluate, for example, the scalability and adaptability of software to evolving hardware, like GPUs, ensuring robust performance and easier maintenance.

The tools are based on the Git repository of the codebase to be able to successfully retrieve all the elements necessary. Tools like Anubis evaluate code complexity and maintenance needs, while Maraudersmap visualizes developer interactions to enhance collaboration. Case studies on eight major CFD codes reveal strategies to streamline code evolution, improve scalability, and reduce maintenance costs, helping to align software with contemporary computational demands.

An access to the Git Repository is mandatory to evaluate the codebase.

Having an interaction in order to discard parts of lesser importance (e.g. code automatically generated,or  deprecated features)

Tool license

MIT Licence

Tool URL

https://pypi.org/project/anubis/

https://pypi.org/project/maraudersmap/

Programming language

Python

Availability

Public