Home > Programs > Zenith Grants > Zenith Grant Awardee

Zenith Grant Awardee

Dr. Gaurav Khanna

University of Massachusetts at Dartmouth

Project Title

Numerical Techniques for Solving Models of Quantum Gravity

Project Summary

A complete theory of quantum gravity is needed to answer some of the most foundational questions in science: How did our universe begin? What is the nature of space and time? What are "singularities''? Loop quantum cosmology is an application of loop quantum gravity, a theory that leads to space-time that is discrete at Planck scales (about thirty-three orders of magnitude smaller than a centimeter) to models in cosmology. One important result of LQC is that it is free of "singularities" – such as the "big bang", replacing it by a "big bounce" – and thus enabling us to theoretically probe the events that lead to the beginning of our universe. This project is about understanding and analyzing a specific aspect of LQC (the semi-classical limit) using numerical computation. What is particularly unique about our proposed research is that we intend to use the Sony Playstation 3 for this computation because it is particularly well suited for this work. The computation would have otherwise required a moderate supercomputing level effort. Thus, this project will also explore the possibility of obtaining an extremely high "bang-per-buck" for scientific computation using the PS3, compared with more traditional systems.

Technical Abstract

Loop quantum cosmology is a symmetry-reduced application of loop quantum gravity, a theory that leads to space-time that is discrete at Planck scales. One important result of LQC is that it is free of singularities. This is seen as a general consequence of the quantum evolution equation, which is a "difference equation" for the wave function and does not break down where the classical singularity is located. In addition, LQC has led to several other interesting directions for research.

Since LQC involves discrete, difference equations, this means that these models are particularly suited for study using numerical analysis and computation. In this proposal, we will develop numerical techniques for solving generic, "lattice-refined" models in LQC. We expect these techniques would require moderate supercomputing level resources, and we propose a unique approach to achieve that – the Sony Playstation 3 (PS3). The PS3's processor is particularly suited for this type of computation and is likely to yield an extremely high "performance-per-dollar" for such numerical computation, compared with traditional systems. Thus, this project's outcome will also have a strong impact on scientific computation, at large.

Skip to content