arXiv subjects

Thomas Lippert

Publications and source records attributed to Thomas Lippert.

54 records · Page 3Linked to original sources

arXiv2010

Scaling study for 2 HEX smeared fermions: hadron and quark masses

The goal of this study is to investigate the scaling behaviour of our 2 HEX action. For this purpose, we compute the $N_f=3$ spectrum and compare the results to our 6 EXP action. We find a large scaling window up to $\sim 0.15\,\mathrm{fm}$ along with small scaling corrections at the 2%-level and full compatibility with our previous study. As a second important observable to be tested for scaling, we chose the non-perturbatively renormalized quenched strange quark mass. Here we find a fairly flat scaling with a broad scaling range up to $\simeq 0.15\,\mathrm{fm}$ and perfect agreement with the literature.

hep-lat↗

arXiv2008

Topological tunneling with Dynamical overlap fermions

Tunneling between different topological sectors with dynamical chiral fermions is difficult because of a poor mass scaling of the pseudo-fermion estimate of the determinant. For small fermion masses it is virtually impossible using standard methods. However, by projecting out the small Wilson eigenvectors from the overlap operator, and treating the correction determinant exactly, we can significantly increase the rate of topological sector tunneling and reduce substantially the auto-correlation time. We present and compare a number of different approaches, and advocate a method which allows topological tunneling even at low mass with little addition to the computational cost.

hep-lat↗

arXiv2005

String breaking

We numerically investigate the transition of the static quark-antiquark string into a static-light meson-antimeson system. Improving noise reduction techniques, we are able to resolve the signature of string breaking dynamics for Nf=2 lattice QCD at zero temperature. We discuss the lattice techniques used and present results on energy levels and mixing angle of the static two-state system. We visualize the action density distribution in the region of string breaking as a function of the static colour source-antisource separation. The results can be related to properties of quarkonium systems.

hep-lat↗

arXiv2005

Anatomy of String Breaking in QCD

We investigate the string breaking mechanism in n_f=2 QCD. We discuss the lattice techniques used and present results on energy levels and mixing angle of the static BBbar|QbarQ two-state system. The string breaking is visualized, by means of an animation of the action density distribution as a function of the static colour source-antisource separation.

hep-lat↗

arXiv2005

Basics of Modelling the Pedestrian Flow

For the modelling of pedestrian dynamics we treat persons as self-driven objects moving in a continuous space. On the basis of a modified social force model we qualitatively analyze the influence of various approaches for the interaction between the pedestrians on the resulting velocity-density relation. To focus on the role of the required space and remote force we choose a one-dimensional model for this investigation. For those densities, where in two dimensions also passing is no longer possible and the mean value of the velocity depends primarily on the interaction, we obtain the following result: If the model increases the required space of a person with increasing current velocity, the reproduction of the typical form of the fundamental diagram is possible. Furthermore we demonstrate the influence of the remote force on the velocity-density relation.

physics.soc-ph↗

arXiv2005

Observation of String Breaking in QCD

hep-lat↗

arXiv2004

Numerical Methods for the QCD Overlap Operator:III. Nested Iterations

The numerical and computational aspects of chiral fermions in lattice quantum chromodynamics are extremely demanding. In the overlap framework, the computation of the fermion propagator leads to a nested iteration where the matrix vector multiplications in each step of an outer iteration have to be accomplished by an inner iteration; the latter approximates the product of the sign function of the hermitian Wilson fermion matrix with a vector. In this paper we investigate aspects of this nested paradigm. We examine several Krylov subspace methods to be used as an outer iteration for both propagator computations and the Hybrid Monte-Carlo scheme. We establish criteria on the accuracy of the inner iteration which allow to preserve an a priori given precision for the overall computation. It will turn out that the accuracy of the sign function can be relaxed as the outer iteration proceeds. Furthermore, we consider preconditioning strategies, where the preconditioner is built upon an inaccurate approximation to the sign function. Relaxation combined with preconditioning allows for considerable savings in computational efforts up to a factor of 4 as our numerical experiments illustrate. We also discuss the possibility of projecting the squared overlap operator into one chiral sector.

hep-lat↗

arXiv2003

Recent development and perspectives of machines for lattice QCD

I highlight recent progress in cluster computer technology and assess status and prospects of cluster computers for lattice QCD with respect to the development of QCDOC and apeNEXT. Taking the LatFor test case, I specify a 512-processor QCD-cluster better than 1$/Mflops.

hep-lat↗

arXiv2003

Moments of Nucleon Generalized Parton Distributions in Lattice QCD

Calculation of moments of generalized parton distributions in lattice QCD requires more powerful techniques than those previously used to calculate moments of structure functions. Hence, we present a novel approach that exploits the full information content from a given lattice configuration by measuring an overdetermined set of lattice observables to provide maximal statistical constraints on the generalized form factors at a given virtuality, t. In an exploratory investigation using unquenched QCD configurations at intermediate sea quark masses, we demonstrate that our new technique is superior to conventional methods and leads to reliable numerical signals for the n=2 flavor singlet generalized form factors up to 3 GeV^2. The contribution from connected diagrams in the flavor singlet sector to the total quark angular momentum is measured to an accuracy of the order of one percent.

hep-lat↗

arXiv2003

Fast Parallel I/O on Cluster Computers

Today's cluster computers suffer from slow I/O, which slows down I/O-intensive applications. We show that fast disk I/O can be achieved by operating a parallel file system over fast networks such as Myrinet or Gigabit Ethernet. In this paper, we demonstrate how the ParaStation3 communication system helps speed-up the performance of parallel I/O on clusters using the open source parallel virtual file system (PVFS) as testbed and production system. We will describe the set-up of PVFS on the Alpha-Linux-Cluster-Engine (ALiCE) located at Wuppertal University, Germany. Benchmarks on ALiCE achieve write-performances of up to 1 GB/s from a 32-processor compute-partition to a 32-processor PVFS I/O-partition, outperforming known benchmark results for PVFS on the same network by more than a factor of 2. Read-performance from buffer-cache reaches up to 2.2 GB/s. Our benchmarks are giant, I/O-intensive eigenmode problems from lattice quantum chromodynamics, demonstrating stability and performance of PVFS over Parastation in large-scale production runs.

cs.DC↗

arXiv2002

Cost of QCD simulations with n_f=2 dynamical Wilson fermions

Cost estimates for simulations of full QCD with n_f=2 Wilson fermions by hybrid Monte Carlo are presented. The extrapolations are based on the average number of iterations of the iterative solver within the fermionic part of the HMC molecular dynamics, which is closely related to the minimal eigenvalue of M^+M The cost formula is determined as a product of the scaling functions of iterative solver and integrated autocorrelation time as function of the inverse lattice pseudoscalar mass. Timings by SESAM/TxL allow to fix the pre-factor. It is demonstrated that a 2-flavor dynamical determination of light hadron masses with a statistical precision comparable to the corresponding quenched results from CP-PACS can be an appropriate task for a 100 Tflops system.

hep-lat↗

arXiv2001

Quark mass effects on the topological susceptibility in QCD

We investigate topology in lattice simulations of QCD with two flavours of dynamical Wilson fermions. At various sea quark masses we find reasonable agreement between results for the topological charge from fermionic and gluonic definitions, the latter with cooling. We study correlations of the topological charge with light hadronic observables and find effects in the flavour singlet pseudoscalar channel. The quark mass dependence of the topological susceptibility, $χ$, is in agreement with the leading order small m_pi expectation, chi=(f_pi m_pi)^2/4.

hep-lat↗

arXiv2000

One-Flavour Hybrid Monte Carlo with Wilson Fermions

The Wilson fermion determinant can be written as product of the determinants of two hermitian positive definite matrices. This formulation allows to simulate non-degenerate quark flavors by means of the hybrid Monte Carlo algorithm. A major numerical difficulty is the occurrence of nested inversions. We construct a Uzawa iteration scheme which treats the nested system within one iterative process.

hep-lat↗

arXiv2000

Static potentials and glueball masses from QCD simulations with Wilson sea quarks

We calculate glueball and torelon masses as well as the lowest lying hybrid potential in addition to the static ground state potential in lattice simulations of QCD with two flavours of dynamical Wilson fermions. The results are obtained on lattices with $16^3\times 32$ and $24^3\times 40$ sites at $β=5.6$, corresponding to a lattice spacing, $a^{-1}=2.65^{+5}_{-8}$ GeV, as determined from the Sommer force radius, at physical sea quark mass. The range spanned in the present study of five different quark masses is reflected in the ratios, $0.83\geq m_π/m_ρ\geq 0.57$.

hep-lat↗

arXiv1998

Hyper-Systolic Matrix Multiplication

A novel parallel algorithm for matrix multiplication is presented. The hyper-systolic algorithm makes use of a one-dimensional processor abstraction. The procedure can be implemented on all types of parallel systems. It can handle matrix-vector multiplications as well as transposed matrix products.

cs.MS↗

arXiv1997

The Hybrid Monte Carlo Algorithm for Quantum Chromodynamics

The Hybrid Monte Carlo (HMC) algorithm currently is the favorite scheme to simulate quantum chromodynamics including dynamical fermions. In this talk-which is intended for a non-expert audience--I want to bring together methodical and practical aspects of the HMC for full QCD simulations. I will comment on its merits and shortcomings, touch recent improvements and try to forecast its efficiency and rôle in future full QCD simulations.

hep-lat↗

arXiv1997

FFT for the APE Parallel Computer

We present a parallel FFT algorithm for SIMD systems following the `Transpose Algorithm' approach. The method is based on the assignment of the data field onto a 1-dimensional ring of systolic cells. The systolic array can be universally mapped onto any parallel system. In particular for systems with next-neighbour connectivity our method has the potential to improve the efficiency of matrix transposition by use of hyper-systolic communication. We have realized a scalable parallel FFT on the APE100/Quadrics massively parallel computer, where our implementation is part of a 2-dimensional hydrodynamics code for turbulence studies. A possible generalization to 4-dimensional FFT is presented, having in mind QCD applications.

hep-lat↗

arXiv1995

Many Masses on One Stroke: Economic Computation of Quark Propagators

The computational effort in the calculation of Wilson fermion quark propagators in Lattice Quantum Chromodynamics can be considerably reduced by exploiting the Wilson fermion matrix structure in inversion algorithms based on the non-symmetric Lanczos process. We consider two such methods: QMR (quasi minimal residual) and BCG (biconjugate gradients). Based on the decomposition $M/κ={\bf 1}/κ-D$ of the Wilson mass matrix, using QMR, one can carry out inversions on a {\em whole} trajectory of masses simultaneously, merely at the computational expense of a single propagator computation. In other words, one has to compute the propagator corresponding to the lightest mass only, while all the heavier masses are given for free, at the price of extra storage. Moreover, the symmetry $γ_5\, M= M^{\dagger}\,γ_5$ can be used to cut the computational effort in QMR and BCG by a factor of two. We show that both methods then become---in the critical regime of small quark masses---competitive to BiCGStab and significantly better than the standard MR method, with optimal relaxation factor, and CG as applied to the normal equations.

hep-lat↗