This video presents the design of a parallel algorithm for LU decomposition, formulating computation supersteps based on a chosen data distribution and inserting communication supersteps where needed. The BSP cost is analysed and as a result the distribution choice is first narrowed down to Cartesian, then to 2D cyclic, and finally to square.
This video corresponds to Section 2.3 of the book Parallel Scientific Computation: A Structured Approach Using BSP, Second Edition, by Rob H. Bisseling, Oxford University Press, 2020.
An expanded set of slides, solutions to the homework questions, and software accompanying the book can all be found on my personal book page:
https://webspace.science.uu.nl/~bisse101/Book2/psc2
