A. Schoneveld

Self-organized criticality in simulated correlated systems

In this paper we study the influence of spatio-temporal correlations on the dynamic runtime behavior of the optimistic parallel Time Warp simulation method. By means of Ising spin simulation, we show experimentally that the probability distribution of the number of rolled back events behaves as a power-law distribution over a large range of sub-critical Ising temperatures and decays exponentially for super-critical Ising temperatures. The experimental results indicate that for critical Ising temperatures, where long-range correlations occur, the computational complexity of Time Warp and physical complexity of the Ising spin model are entangled and contribute both to the runtime behavior in a nonlinear way.

Load balancing by redundant decomposition and mapping

Abstract In this paper a new methodology for load balancing parallel processes on parallel systems is proposed. The problem of load balancing is considered to be an NP-hard optimization task. Taking static parallel finite element applications as a case study, the benefits and losses that follow from applying the methodology are studied. It is found that the proposed methodology can be especially useful for load balancing in asymmetric processor topologies, and therefore is of importance for work load balancing in workstation clusters.

Load Balancing by Redundant Decomposition and Mapping

In this paper a methodology is presented that has been developed in the CAMAS3 project for the purpose of decomposition and mapping of parallel processes to processor topologies. The methodology has been implemented in terms of a toolset, thus allowing automatic decomposition and mapping of parallel processes. The parallel processes and processors are modelled according to a generally applicable formalism, based on the so-called virtual particle model. As a case study the presented methodology is applied to parallel finite element simulations.

P-CAM: a framework for parallel complex systems simulations

Abstract History has taught that the design and implementation of an efficient parallel simulation program is a tedious and error prone process. Methods that can circumvent the parallelization steps in this process are usually warmly welcomed by parallel simulation architects. In this paper we introduce a Parallel Cellular Automata Modeling environment (P-CAM) for doing spatial load balancing on arbitrary connected grids or task graphs. This environment adopts an object oriented application framework in which we can instantiate a variety of simulation problems. We have implemented a kernel, based on this framework, which facilitates dynamic load balancing and supports process migration and irregular interprocess communication patterns. The design of the kernel enables a transparent implementation of complex systems models onto arbitrary parallel computer systems. We show that the design of a parallel simulation program can be assisted by using the P-CAM kernel.

Task Allocation by Parallel Evolutionary Computing

In this paper we will investigate the applicability of parallel evolutionary algorithms to the task allocation problem?a long standing problem in parallel computing. Three different evolutionary optimization strategies, genetic algorithms, simulated annealing, and steepest descent, are formulated in a parallel generic framework. In order to enhance the performance of the strategies, a number of adjustments that exploit problem specific knowledge is proposed. We adopt a parametric description of static parallel applications. As a consequence, a theoretical analysis of the task allocation solution space can be conducted with a method originating from computational biology. The prediction following from this analysis, i.e., simulated annealing performs optimally on the solution space, is supported by experimental results.

Spatio-temporal correlations and rollback distributions in optimistic simulations

In this paper we study the influence of spatio-temporal correlations on the dynamic runtime behavior of the optimistic parallel Time Warp simulation method. By using the Ising spin model, we show experimentally that the distribution of the number of rolled back events behaves as a power-law distribution over a large range of sub-critical Ising temperatures and decays exponentially or super-critical Ising temperatures. For critical Ising temperatures, where long-range correlations occur the computational complexity, of Time Warp and physical complexity of the Ising spin model are entangled and contribute both to the runtime behavior in a nonlinear way.

Preserving Locality for Optimal Parallelism in Task Allocation

Genetic Algorithms have been applied to several combinatorial optimisation problems, including the well known task allocation problem, originating from parallel computing. We introduce random task graphs as a model of applications which display irregular global communication patterns. Uniform crossover is the standard genetic recombination operator, that is applied to solution encoded chromosomes. However, application of a uniform crossover may heavily disrupt low cost sub-solutions, or building blocks, of a chromosome. Therefore, we define a locality preserving recombination operator, exploiting the connectivity of the task graph. Experiments show that this new operator increases the convergence rate of the Genetic Algorithm applied to the task allocation problem.

A framework for dynamic load balancing: a case study on explosive containment simulation

Abstract Methods that can circumvent the implementation of parallelization steps are usually warmly welcomed by parallel simulation architects. In this paper we apply the parallel cellular automata modeling environment (P-CAM) to a parallel finite element simulation. P-CAM adopts a particle-based framework in which we can instantiate a variety of simulation problems. However, it can also be used to monitor and steer a parallel application outside this framework. A meta-model of the finite element grid is used within the framework to dynamically repartition the grid during the simulation. Based on this meta model a dynamic load balancing process, implemented in P-CAM, decides on a possible local repartitioning of the mesh. The system is successfully used to simulate highly dynamic transient behavior of an explosive detonating under a water surface.

Dynamic Load Balancing in Parallel Finite Element Simulations

In this paper we introduce a new method for parallelizing Finite Element simulations enabling the use of dynamic load balancing. A physical space partitioning is obtained by dividing the bounding cube into a large number of sub cubes. The cube mesh together with a workload attribute assigned to each cube is used to present an abstract view of the simulation. Based on this abstract view a dynamic load balancing process decides on a possible local repartitioning of the mesh. The dynamic load balancing process itself is diffusion based, that is cubes are migrated between neighboring partitions. A parallel simulation framework (P-CAM) is used to implement the dynamic load balancer.