JOSTLE: multilevel graph partitioning software: an overview

In this chapter we look at JOSTLE, the multilevel graph-partitioning software package, and highlight some of the key research issues that it addresses. We first outline the core algorithms and place it in the context of the multilevel refinement paradigm. We then look at issues relating to its use as a tool for parallel processing and, in particular, partitioning in parallel. Since its first release in 1995, JOSTLE has been used for many mesh-based parallel scientific computing applications and so we also outline some enhancements such as multiphase mesh-partitioning, heterogeneous mapping and partitioning to optimise subdomain shape

Automatic generation of portable multi-dimensionally decomposed parallel structured mesh software

Abstract not available

Modelling of industrial multi-physics processes—a key role for computational mechanics.

In the analysis of industrial processes, there is an increasing emphasis on systems governed by interacting continuum phenomena. Mathematical models of such multi-physics processes can only be achieved for practical simulations through computational solution procedures—computational mechanics. Examples of such multi-physics systems in the context of metals processing are used to explore some of the key issues. Finite-volume methods on unstructured meshes are proposed as a means to achieve efficient rapid solutions to such systems. Issues associated with the software design, the exploitation of high performance computers, and the concept of the virtual computational-mechanics modelling laboratory are also addressed in this context.

PHYSICA - A software environment for the modelling of multi-physics phenomena

A brief description of a software environment in FORTRAN77 for the modelling of multi-physics phenomena is given. The numerical approach is based on finite volume methods but extended to unstructured meshes (ie. FV-UM). A range of interacting solution procedures for turbulent fluid flow, heat transfer with solidification/melting and elasto-visco-plastic solid mechanics are implemented in the first version of PHYSICA, which will be released in source code form to the academic community in late 1995.

A scalable strategy for the parallelization of multiphysics unstructured mesh-iterative codes on distributed-memory systems

Realizing scalable performance on high performance computing systems is not straightforward for single-phenomenon codes (such as computational fluid dynamics [CFD]). This task is magnified considerably when the target software involves the interactions of a range of phenomena that have distinctive solution procedures involving different discretization methods. The problems of addressing the key issues of retaining data integrity and the ordering of the calculation procedures are significant. A strategy for parallelizing this multiphysics family of codes is described for software exploiting finite-volume discretization methods on unstructured meshes using iterative solution procedures. A mesh partitioning-based SPMD approach is used. However, since different variables use distinct discretization schemes, this means that distinct partitions are required; techniques for addressing this issue are described using the mesh-partitioning tool, JOSTLE. In this contribution, the strategy is tested for a variety of test cases under a wide range of conditions (e.g., problem size, number of processors, asynchronous / synchronous communications, etc.) using a variety of strategies for mapping the mesh partition onto the processor topology.

An integrated modeling approach to solder joint formation

The attachment of electronic components to printed circuit boards using solder material is a complex process. This paper presents a novel modeling methodology, which integrates the governing physics taking place. Multiphysics modeling technology, imbedded into the simulation tool—PHYSICA is used to simulate fluid flow, heat transfer, solidification, and stress evolution in an integrated manner. Results using this code are presented, detailing the mechanical response of two solder materials as they cool, solidify and then deform. The shape that a solder joint takes upon melting is predicted using the SURFACE EVOLVER code. Details are given on how these predictions can be used in the PHYSICA code to provide a modeling route by which the shape, solidification history, and resulting stress profiles can be predicted.

Computational issues in the modelling of materials-based manufacturing processes

The manufacture of materials products involves the control of a range of interacting physical phenomena. The material to be used is synthesised and then manipulated into some component form. The structure and properties of the final component are influenced by both interactions of continuum-scale phenomena and those at an atomistic-scale level. Moreover, during the processing phase there are some properties that cannot be measured (typically the liquid-solid phase change). However, it seems there is a potential to derive properties and other features from atomistic-scale simulations that are of key importance at the continuum scale. Some of the issues that need to be resolved in this context focus upon computational techniques and software tools facilitating: (i) the multiphysics modeling at continuum scale; (ii) the interaction and appropriate degrees of coupling between the atomistic through microstructure to continuum scale; and (iii) the exploitation of high-performance parallel computing power delivering simulation results in a practical time period. This paper discusses some of the attempts to address each of the above issues, particularly in the context of materials processing for manufacture.

Bulk synchronous parallelization of industrial electromagentic software

The parallelization of existing/industrial electromagnetic software using the bulk synchronous parallel (BSP) computation model is presented. The software employs the finite element method with a preconditioned conjugate gradient-type solution for the resulting linear systems of equations. A geometric mesh-partitioning approach is applied within the BSP framework for the assembly and solution phases of the finite element computation. This is combined with a nongeometric, data-driven parallel quadrature procedure for the evaluation of right-hand-side terms in applications involving coil fields. A similar parallel decomposition is applied to the parallel calculation of electron beam trajectories required for the design of tube devices. The BSP parallelization approach adopted is fully portable, conceptually simple, and cost-effective, and it can be applied to a wide range of finite element applications not necessarily related to electromagnetics.

The integration of an intelligent knowledge-based system into engineering software using the blackboard structure

Over recent years there has been an increase in the use of generic Computational Fluid Dynamics (CFD) software packages spread across various application fields. This has created the need for the integration of expertise into CFD software. Expertise can be integrated into CFD software in the form of an Intelligent Knowledge-Based System (IKBS). The advantages of integrating intelligence into generic engineering software are discussed with a special view to software engineering considerations. The software modelling cycle of a typical engineering problem is identified and the respective expertise and user control needed for each modelling phase is shown. The requirements of an IKBS for CFD software are discussed and compared to current practice. The blackboard software architecture is presented. This is shown to be appropriate for the integration of an IKBS into an engineering software package. This is demonstrated through the presentation of the prototype CFD software package FLOWES.

Estimating the flow rate capacity of an overturned rail carriage end exit in the presence of smoke

While incidents requiring the rapid egress of passengers from trains are infrequent, perhaps the most challenging scenario for passengers involves the evacuation from an overturned carriage subjected to fire. In this paper we attempt to estimate the flow rate capacity of an overturned rail carriage end exit. This was achieved through two full-scale evacuation experiments, in one of which the participants were subjected to non-toxic smoke. The experiments were conducted as part of a pilot study into evacuation from rail carriages. In reviewing the experimental results, it should be noted that only a single run of each trial was undertaken with a limited — though varied — population. As a result it is not possible to test the statistical significance of the evacuation times quoted and so the results should be treated as indicative rather than definitive. The carriage used in the experiments was a standard class Mark IID which, while an old carriage design, shares many features with those carriages commonly found on the British rail network. In the evacuation involving smoke, the carriage end exit was found to achieve an average flow rate capacity of approximately 5.0 persons/min. The average flow rate capacity of the exit without smoke was found to be approximately 9.2 persons/min. It was noted that the presence of smoke tended to reduce significantly the exit flow rate. Due to the nature of the experimental conditions, these flow rates are considered optimistic. Finally, the authors make several recommendations for improving survivability in rail accidents. Copyright © 2000 John Wiley & Sons, Ltd.

Multilevel mesh partitioning for heterogeneous communication networks

Multilevel algorithms are a successful class of optimisation techniques which address the mesh partitioning problem for distributing unstructured meshes onto parallel computers. They usually combine a graph contraction algorithm together with a local optimisation method which refines the partition at each graph level. To date these algorithms have been used almost exclusively to minimise the cut edge weight in the graph with the aim of minimising the parallel communication overhead, but recently there has been a perceived need to take into account the communications network of the parallel machine. For example the increasing use of SMP clusters (systems of multiprocessor compute nodes with very fast intra-node communications but relatively slow inter-node networks) suggest the use of hierarchical network models. Indeed this requirement is exacerbated in the early experiments with meta-computers (multiple supercomputers combined together, in extreme cases over inter-continental networks). In this paper therefore, we modify a multilevel algorithm in order to minimise a cost function based on a model of the communications network. Several network models and variants of the algorithm are tested and we establish that it is possible to successfully guide the optimisation to reflect the chosen architecture.

Utilising computational fluid dynamics (CFD) for the modelling of granular material in large-scale engineering processes

In this paper, the framework is described for the modelling of granular material by employing Computational Fluid Dynamics (CFD). This is achieved through the use and implementation in the continuum theory of constitutive relations, which are derived in a granular dynamics framework and parametrise particle interactions that occur at the micro-scale level. The simulation of a process often met in bulk solids handling industrial plants involving granular matter, (i.e. filling of a flat-bottomed bin with a binary material mixture through pneumatic conveying-emptying of the bin in core flow mode-pneumatic conveying of the material coming out of a the bin) is presented. The results of the presented simulation demonstrate the capability of the numerical model to represent successfully key granular processes (i.e. segregation/degradation), the prediction of which is of great importance in the process engineering industry.