Solving Scalar Fields in AdS Space-Time using Parallel Computing and GPUs

Abstract

Computers have been around a long time and used for many resources like gaming, emailing and science. Concerning the sciences, computers are excellent at repetitive tasks and crunching numbers without (much) error. In data science there are two methods of programming - serial and parallel. Both have the same goal and endpoint however depending on the problem, one might be more efficient in solving the problem. My thesis is on using parallel computing with GPUs to numerically solve a system of non-linear first order differential equations that describe scalar fields in AdS space-time. Since this has been done by another using CPUs, it will be important to compare that the simple case matches that of the previous paper. However once we have confirmed that our results match the previously found results, we can use graphic processing units to solve the problem to higher order and more accuracy.