This project makes use of a novel approach for numerical lattice simulations of the strong nuclear force. Such simulations support an extensive experimental program in the search for new physics, and, as these searches continue, increased precision is required.
Our approach of master field simulations enables larger volumes and finer lattice spacings, crucial for next-generation precision.
The novelty is to employ a smaller number of significantly larger-volume quantum gauge fields, using spatial averaging of local observables.
Accumulating statistics in this manner circumvents the infamous topology-freezing problem of conventional simulations and can further reduce the critical slowing down of algorithms near the continuum limit. With the 120 Mch requested here, we will generate data needed for the first master field calculations with varying lattice spacing.
The fields will enable us to calculate the neutron electric dipole moment, charm-to-light semileptonic decays, and the inclusive rate R(e^+ e^- →hadrons), each of which profit from the master field approach and are of direct importance for new physics searches.
The approach is uniquely suited to exploit the full potential of large-scale HPC facilities as the huge problem size allows for tuning of the computational density to mask network communication and achieve excellent scaling performance.