Linear Algebra in the Browser powered by WebAssembly
Archit Bhonsle, Ved Patil, Tanvi Valkunde, Trupti Lotlikar · 2022 International Conference for Advancement in Technology (ICONAT) · 2022
Linear algebra lies at the heart of high-performance computing with applications ranging from image processing to machine learning. Libraries like NumPy provide efficient interfaces to work with vectors and matrices and have spurred a revolution in the field of computer science. With over 4 billion worldwide active users, the web is the most prolific platform for people to interact with computers. Developers utilize JavaScript to create dynamic user interfaces that power the web as we know it. Although there have been changes to JavaScript or how JavaScript is “interpreted” by the browser, the performance still doesn't match compiled native code. This paper aims to focus on, WebAssembly, one of the mechanisms to overcome this limitation and write a linear algebra library that will empower developers in creating user experiences that go beyond the traditional web.