Parallel EFG algorithm for heat transfer problems

The main issues when supporting fault tolerance based on checkpointing and rollback recovery for High-Performance applications are related to the scalability of the introduced support, the possibility of analyzing the induced overhead and, in more general terms, the optimization of the trade-off between failure-free and recovery performances. In this paper we describe our contribution in fault tolerance for high-level structured parallelism models. We take a different viewpoint w.r.t. existing contributions, by introducing a methodology to derive interesting properties to support fault tolerance. We show how to apply this methodology to a general data parallel model, deriving useful properties to introduce a class of checkpointing protocols. Thanks to this methodology, this class of protocols is not affected by the described issues. We exemplify two checkpointing protocols and the related rollback recovery techniques. For each protocol we also derive cost models statically describing the failure-free performance, which can be used for performance tuning or to target some Quality of Service parameter. To assess the innovation of the results we analytically and experimentally compare the introduced protocols with two literature protocols. Results show that while the protocols introduced in this paper permit the definition of cost models and have a good scalability, the literature protocols do not always have these properties

show abstract

“…for heat transfer applications [13]); the re-solution of tridiagonal systems (e.g. [14]); computation and communication patterns (e.g.…”

Section: Introductionmentioning

confidence: 99%

Fault tolerance for data parallel programs

Bertolli

Vanneschi

2010

Concurrency and Computation

View full text Add to dashboard Cite

show abstract

“…Some meshless methods have been recently applied to solve transient heat conduction problems. Singh and his coworkers [15,16] used the element-free Galerkin (EFG) method on steady, unsteady and nonlinear heat transfer problems. However, the EFG method has drawbacks; it needs background cells to integrate a weak form over the entire domain or boundary of the problem.…”

mentioning

confidence: 99%

A local Petrov-Galerkin approach with moving Kriging interpolation for solving transient heat conduction problems

Chen

Liew

2010

Comput Mech

View full text Add to dashboard Cite

A meshless Local Petrov-Galerkin approach based on the moving Kriging interpolation (Local Kriging method; LoKriging hereafter) is employed for solving partial different equations that govern the heat flow in two-and three-dimensional spaces. The method is developed based on the moving Kriging interpolation for constructing shape functions at scattered points, and the Heaviside step function is used as a test function in each sub-domain to avoid the need for domain integral in symmetric weak form. As the shape functions possess the Kronecker delta function property, essential boundary conditions can be implemented without any difficulties. The traditional two-point difference method is selected for the time discretization scheme. For computation of 3D problems, a novel local sub-domain from the polyhedrons is used for evaluating the integrals. Several selected numerical examples are presented to illustrate the performance of the LoKriging method.

show abstract

“…Singh and his co-workers [25][26][27][28][29] have done much work on steady, unsteady, nonlinear heat transfer problems based on EFG. Arefmanesh et al [30] and Batra et al [31] put forward a meshless method for solving heat conduction problem based on MLPG.…”

mentioning

confidence: 99%

The reproducing kernel particle method for two-dimensional unsteady heat conduction problems

Cheng

Liew

2009

Comput Mech

View full text Add to dashboard Cite

The numerical solution to two-dimensional unsteady heat conduction problem is obtained using the reproducing kernel particle method (RKPM). A variational method is employed to furnish the discrete equations, and the essential boundary conditions are enforced by the penalty method. Convergence analysis and error estimation are discussed. Compared with the numerical methods based on mesh, the RKPM needs only the scattered nodes instead of meshing the domain of the problem. The effectiveness of the RKPM for two-dimensional unsteady heat conduction problems is examined by two numerical examples.

show abstract

Parallel EFG algorithm for heat transfer problems

Cited by 22 publications

References 34 publications

Fault tolerance for data parallel programs

Fault tolerance for data parallel programs

A local Petrov-Galerkin approach with moving Kriging interpolation for solving transient heat conduction problems

The reproducing kernel particle method for two-dimensional unsteady heat conduction problems

Contact Info

Product

Resources

About