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, Nit, 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 ite...

This paper presents a collocation method with an iterative linear system solver to compute periodic solutions of a system of autonomous delay differential equations (DDEs). We show that the linearized collocation system is equivalent to a discretization of the linearized periodic boundary value problem (BVP). This linear BVP is solved using the Newton-Picard single shooting method ([Int. J. Bif...

The canonical problem of solving a system of linear equations arises in numerous contexts in information theory, communication theory, and related fields. In this contribution, we develop a solution based upon Gaussian belief propagation (GaBP) that does not involve direct matrix inversion. The iterative nature of our approach allows for a distributed message-passing implementation of the solut...

In this work, the HLLC Riemann solver, which is much more robust, simpler and faster than iterative Riemann solvers, is extended to obtain interface conditions in sharp-interface methods for compressible multi-fluid flows. For interactions with general equations of state and material interfaces, a new generalized Roe average is proposed. For single-phase interactions, this new Roe average does ...

Traditional large sparse linear solvers are not suited in a grid computing environment as they require a large amount of synchronization and communication penalizing the performance on this architecture. This paper presents some features of the solver designed during the current GREMLINS (GRid Efficient Method for LINear Systems) project. The GREMLINS solver limits the amount of communication a...

This paper is devoted to a class of iterative methods for solving nonsymmetric or indeenite problems that are dominated by some SPD (symmetric positive deenite) problems. The algorithm is based on a direct solver for the original equation restricted on a small subspace and a given iterative method for the SPD equation. It is shown that any convergent iterative method for the SPD problem will gi...

1. Preconditioned Iterative Solvers with Multicoloring In the previous work [1], author developed an efficient parallel iterative solver for finite-element applications on the Earth Simulator (ES) [2] using multi-level hybrid parallel programming model with MPI and OpenMP. The method employs three-level hybrid parallel programming model for SMP cluster architectures, consisting of MPI, OpenMP a...

Mineral scale deposition in near wellbore regions of injection wells is one of the main challengeable issues during the water injection process, which magnifies the importance of robust models in predicting the amount of mineral scale deposition such as calcium sulfate. One of the main challenges of CaSO4 scale is in carbonated reservoirs, in which sensitive behavior is observed in related to t...

We present PARDISO, a new scalable parallel sparse direct linear solver on shared memory multiprocessors. In this paper, we describe the parallel fac-torization algorithm which utilizes the supernode structure of the matrix to reduce the number of memory references with Level 3 BLAS. We also propose enhancements that signiicantly reduce the communication rate for pipelining parallelism. The res...

Several recovery techniques for parallel iterative methods are presented. First, the implementation of checkpoints in parallel iterative methods is described and analyzed. Then, a simple checkpoint-free faulttolerant scheme for parallel iterative methods, the lossy approach, is presented. When one processor fails and all its data is lost, the system is recovered by computing a new approximate s...