Rayleigh Quotient Iteration with a Multigrid in Energy Preconditioner for Massively Parallel Neutron Transport

TL;DR

This paper demonstrates a novel combination of multigroup Krylov solver, Rayleigh quotient iteration, and multigrid in energy preconditioning in the Denovo code, significantly improving the efficiency and scalability of neutron transport calculations on leadership-class computers.

Contribution

First effective integration of these three methods enabling faster and more scalable eigenvalue calculations in neutron transport simulations.

Findings

RQI converges in fewer iterations than power iteration for large problems.

The combined methods reduce computation time significantly.

Scales efficiently to nearly 276,000 cores.

Abstract

Three complementary methods have been implemented in the code Denovo that accelerate neutral particle transport calculations with methods that use leadership-class computers fully and effectively: a multigroup block (MG) Krylov solver, a Rayleigh quotient iteration (RQI) eigenvalue solver, and a multigrid in energy preconditioner. The multigroup Krylov solver converges more quickly than Gauss Seidel and enables energy decomposition such that Denovo can scale to hundreds of thousands of cores. The new multigrid in energy preconditioner reduces iteration count for many problem types and takes advantage of the new energy decomposition such that it can scale efficiently. These two tools are useful on their own, but together they enable the RQI eigenvalue solver to work. Each individual method has been described before, but this is the first time they have been demonstrated to work together effectively. RQI should converge in fewer iterations than power iteration (PI) for large and challenging problems. RQI creates shifted systems that would not be tractable without the MG Krylov solver. It also creates ill-conditioned matrices that cannot converge without the multigrid in energy preconditioner. Using these methods together, RQI converged in fewer iterations and in less time than all PI calculations for a full pressurized water reactor core. It also scaled reasonably well out to 275,968 cores.