Abstract
QUEST is a part of the SciDAC project on next generation multi-scale quantum
simulation software for strongly correlated materials. It is a Fortran 90/95 package that
implements the determinant quantum Monte Carlo (DQMC) method for simulation of
magnetic, superconducting, and metal-insulator transitions in model Hamiltonians. In this
paper, we show how QUEST is capable of treating lattices of unprecedentedly large sizes
and how this can be fruitful in the study of the physics of trapped fermionic system, in the
development of more efficient solvers for Dynamical Mean Field Theory (DMFT) and as a tool
to test and, in the future, improve diagrammatic approaches such as the Parquet approximation.
We will also present a range of synergistic activities on the development of stable and robust
numerical algorithms and hybrid granularity parallelization scheme that combines algorithmic
and implementation techniques to high-performance DQMC simulation. The work reported
here is a key step forward in achieving the goals of our SciDAC project.
