The Local Bosonic Algorithm applied to the massive Schwinger model

Details The Local Bosonic Algorithm applied to the massive Schwinger model The Local Bosonic Algorithm applied to the massive Schwinger model

2001-03-28

arxiv.org/abs/hep-lat/0103035v1

Physics

High Energy Physics

High Energy Physics - Lattice

106 pages, 54 figures, Ph.D. thesis

Scientific paper

We investigate various variants of the Hermitean version of the Local Bosonic Algorithm proposed by M. L\"uscher. The model used is two-dimensional Quantum Electrodynamics (QED) with two flavours of massive Wilson fermions. The simplicity of the model allows high statistics simulations close to the chiral and continuum limit. To find optimal CPU cost behaviour, we vary the approximation polynomial parameters $n$ and \epsilon$ as well as the number of over-relaxation steps within each trajectory. We find flat behaviour around the optimum and a modest gain with respect to the Hybrid Monte Carlo algorithm for all variants. On the technical side, we demonstrate that a noisy Metropolis acceptance step is possible also for the Hermitean variant. The numerical instabilities appearing in the evaluation Chebyshev polynomial are investigated. We propose a quantitative criterion for these instabilities and a reordering scheme of the roots reducing the problem. The more physical problem of topological charge sectors and metastability is addressed. We find no plateau in the effective pion mass if metastabilities become too large.

Affiliated with

Also associated with

No associations

Rating

The Local Bosonic Algorithm applied to the massive Schwinger model does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.

If you have personal experience with The Local Bosonic Algorithm applied to the massive Schwinger model, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and The Local Bosonic Algorithm applied to the massive Schwinger model will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFWR-SCP-O-659844