Load Balancing Strategies for Parallel SAMR Algorithms
نویسنده
چکیده
Highly resolved solutions of partial differential equations are important in many areas of science and technology nowadays. Only adaptive mesh refinement methods reduce the necessary work sufficiently allowing the calculation of realistic problems. Blockstructured SAMR methods are well-suited for the time-explicit computation of large-scale dynamical problems, but still require parallelization on distributed systems. Unfortunately assigning activities in a load balanced manner is even more difficult than for unstructured AMR applications. Here common partitioning aims for parallel SAMR algorithms are summarized and strategies for mapping the load balancing problem to simpler unstructured meshes are outlined. The research proposed for the participation in SURF 2005 includes the systematization, implementation and quantitative comparison of known modelling and load balancing strategies.
منابع مشابه
A Performance Characterization of Load Balancing Algorithms for Parallel SAMR Applications
We perform a comprehensive performance characterization of load balancing algorithms for parallel structured adaptive mesh refinement (SAMR) applications. Using SAMR, computational resources are dynamically concentrated to areas in need of a high accuracy. Because of the dynamic resource allocation, the workload must repeatedly be partitioned and distributed over the processors. For an efficien...
متن کاملComparison of Dynamic Load Balancing Techniques for a Parallel SAMR Algorithm
Three dynamic load balancing algorithms are considered: a multilevel inverse space lling curve partitioner (MISP), a wavefront diiusion algorithm , and RSB. They are compared for a parallel structured adaptive mesh reenement (SAMR) algorithm based on a blockwise approach. The aim of the study is to sketch a nuanced picture, where a certain type of load balancing algorithm can be shown to be app...
متن کاملDynamic Load Partitioning Strategies for Managing Data of Space and Time Heterogeneity in Parallel SAMR Applications
This paper presents the design and experimental evaluation of two dynamic load partitioning and balancing strategies for parallel Structured Adaptive Mesh Refinement (SAMR) applications: the Level-based Partitioning Algorithm (LPA) and the Hierarchical Partitioning Algorithm (HPA). These techniques specifically address the computational and communication heterogeneity across refinement levels o...
متن کاملDynamic Structured Grid Hierarchy Partitioners Using Inverse Space-Filling Curves
This paper discusses partitioning of dynamic structured grid hierarchies, occuring in structured adaptive mesh re nement (SAMR) applications. When a SAMR method is executed on a parallel computer, the work load will change dynamically. Thus, there is need for dynamic load balancing. Inverse spacelling curve partitioning (ISP) is appealing for load balancing in parallel SAMR, because of its spee...
متن کامل