GPGPU applied to linear algebra for finance: the case of Cholesky decomposition.

Recently, I came across my Master's thesis report from 2010, completed at Cranfield University in the UK as part of a dual degree program with my French engineering school.

At the time, I worked on the Cholesky decomposition implemented on GPGPU (graphics processing unit clusters), a field that was explored for massively parallel computing. This project, in collaboration with the London-based company Excelian and Cranfield University, aimed to improve the performance of Monte Carlo simulations for price forecasting or basket of European options in finance.

I invite you to explore my report in PDF format. Although it was completed almost 15 years ago and most likely obsolete today, this work provides insight into the technical challenges of the time in the field of financial computing acceleration.

Commentaires

jafar Abdullah

This is a fascinating throwback! It's great to see how early work with GPGPU and Cholesky decomposition helped speed up financial simulations. Even if tech has changed, your thesis shows the solid groundwork that shaped modern computing in finance. Thanks for sharing this valuable piece of history and learning!

Add new comment

Your name will be publicly displayed along with your comment.
Your email will be kept private and only used to notify you.
On internet, you can be who you want. Please be someone nice :)