Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userExtrapolation Methods for the Dirac Inverter in Hybrid Monte Carlo
65
0
0.0
(
0
)
Ask ChatGPT about the research
No Arabic abstract
In Hybrid Monte Carlo(HMC) simulations for full QCD, the gauge fields evolve smoothly as a function of Molecular Dynamics (MD) time. Thus we investigate improved methods of estimating the trial solutions to the Dirac propagator as superpositions of the solutions in the recent past. So far our best extrapolation method reduces the number of Conjugate Gradient iterations per unit MD time by about a factor of 4. Further improvements should be forthcoming as we further exploit the information of past trajectories.
rate research
Read More
The extreme computational costs of calculating the sign of the Wilson matrix within the overlap operator have so far prevented four dimensional dynamical overlap simulations on realistic lattice sizes, because the computational power required to invert the overlap operator, the time consuming part of the Hybrid Monte Carlo algorithm, is too high. In this series of papers we introduced the optimal approximation of the sign function and have been developing preconditioning and relaxation techniques which reduce the time needed for the inversion of the overlap operator by over a factor of four, bringing the simulation of dynamical overlap fermions on medium-size lattices within the range of Teraflop-computers. In this paper we adapt the HMC algorithm to overlap fermions. We approximate the matrix sign function using the Zolotarev rational approximation, treating the smallest eigenvalues of the Wilson operator exactly within the fermionic force. We then derive the fermionic force for the overlap operator, elaborating on the problem of Dirac delta-function terms from zero crossings of eigenvalues of the Wilson operator. The crossing scheme proposed shows energy violations which are better than O($Deltatau^2$) and thus are comparable with the violations of the standard leapfrog algorithm over the course of a trajectory. We explicitly prove that our algorithm satisfies reversibility and area conservation. Finally, we test our algorithm on small $4^4$, $6^4$, and $8^4$ lattices at large masses.
We investigate reversibility violations in the Hybrid Monte Carlo algorithm. Those violations are inevitable when computers with finite numerical precision are being used. In SU(2) gauge theory, we study the dependence of observables on the size of the reversibility violations. While we cannot find any statistically significant deviation in observables related to the simulated physical model, algorithmic specific observables signal an upper bound for reversibility violations below which simulations appear unproblematic. This empirically derived condition is independent of problem size and parameter values, at least in the range of parameters studied here.
UKQCDs dynamical fermion project uses the Generalised Hybrid Monte-Carlo (GHMC) algorithm to generate QCD gauge configurations for a non-perturbatively O(a) improved Wilson action with two degenerate sea-quark flavours. We describe our implementation of the algorithm on the Cray-T3E, concentrating on issues arising from code verification and performance optimisation, such as parameter tuning, reversibility, the effect of precision, the choice of matrix inverter and the behaviour of different molecular dynamics integration schemes.
In these lecture notes some applications of Monte Carlo integration methods in Quantum Field Theory - in particular in Quantum Chromodynamics - are introduced and discussed.
Hasenbusch has proposed splitting the pseudo-fermionic action into two parts, in order to speed-up Hybrid Monte Carlo simulations of QCD. We have tested a different splitting, also using clover-improved Wilson fermions. An additional speed-up between 5 and 20% over the original proposal was achieved in production runs.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
