Generating OpenMP code using an interactive parallelization environment

Code parallelization using OpenMP for shared memory systems is relatively easier than using message passing for distributed memory systems. Despite this, it is still a challenge to use OpenMP to parallelize application codes in a way that yields effective scalable performance when executed on a shared memory parallel system. We describe an environment that will assist the programmer in the various tasks of code parallelization and this is achieved in a greatly reduced time frame and level of skill required. The parallelization environment includes a number of tools that address the main tasks of parallelism detection, OpenMP source code generation, debugging and optimization. These tools include a high quality, fully interprocedural dependence analysis with user interaction capabilities to facilitate the generation of efficient parallel code, an automatic relative debugging tool to identify erroneous user decisions in that interaction and also performance profiling to identify bottlenecks. Finally, experiences of parallelizing some NASA application codes are presented to illustrate some of the benefits of using the evolving environment.

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.

Effects of structure on the comprehensibility of formal specifications

Use of structuring mechanisms (such as modularisation) is widely believed to be one of the key ways to improve software quality. Structuring is considered to be at least as important for specification documents as for source code, since it is assumed to improve comprehensibility. Yet, as with most widely held assumptions in software engineering, there is little empirical evidence to support this hypothesis. Also, even if structuring can be shown to he a good thing, we do not know how much structuring is somehow optimal. One of the more popular formal specification languages, Z, encourages structuring through its schema calculus. A controlled experiment is described in which two hypotheses about the effects of structure on the comprehensibility of Z specifications are tested. Evidence was found that structuring a specification into schemas of about 20 lines long significantly improved comprehensibility over a monolithic specification. However, there seems to be no perceived advantage in breaking down the schemas into much smaller components. The experiment can he fully replicated.

Three parallel programming paradigms: Comparisons on an archetypal PDE computation

Three paradigms for distributed-memory parallel computation that free the application programmer from the details of message passing are compared for an archetypal structured scientific computation -- a nonlinear, structured-grid partial differential equation boundary value problem -- using the same algorithm on the same hardware. All of the paradigms -- parallel languages represented by the Portland Group's HPF, (semi-)automated serial-to-parallel source-to-source translation represented by CAP-Tools from the University of Greenwich, and parallel libraries represented by Argonne's PETSc -- are found to be easy to use for this problem class, and all are reasonably effective in exploiting concurrency after a short learning curve. The level of involvement required by the application programmer under any paradigm includes specification of the data partitioning, corresponding to a geometrically simple decomposition of the domain of the PDE. Programming in SPMD style for the PETSc library requires writing only the routines that discretize the PDE and its Jacobian, managing subdomain-to-processor mappings (affine global-to-local index mappings), and interfacing to library solver routines. Programming for HPF requires a complete sequential implementation of the same algorithm as a starting point, introduction of concurrency through subdomain blocking (a task similar to the index mapping), and modest experimentation with rewriting loops to elucidate to the compiler the latent concurrency. Programming with CAPTools involves feeding the same sequential implementation to the CAPTools interactive parallelization system, and guiding the source-to-source code transformation by responding to various queries about quantities knowable only at runtime. Results representative of "the state of the practice" for a scaled sequence of structured grid problems are given on three of the most important contemporary high-performance platforms: the IBM SP, the SGI Origin 2000, and the CRAYY T3E.

Turbulence modelling and its impact on CFD predictions for cooling of electronic components

This paper will discuss Computational Fluid Dynamics (CFD) results from an investigation into the accuracy of several turbulence models to predict air cooling for electronic packages and systems. Also new transitional turbulence models will be proposed with emphasis on hybrid techniques that use the k-ε model at an appropriate distance away from the wall and suitable models, with wall functions, near wall regions. A major proportion of heat emitted from electronic packages can be extracted by air cooling. This flow of air throughout an electronic system and the heat extracted is highly dependent on the nature of turbulence present in the flow. The use of CFD for such investigations is fast becoming a powerful and almost essential tool for the design, development and optimization of engineering applications. However turbulence models remain a key issue when tackling such flow phenomena. The reliability of CFD analysis depends heavily on the turbulence model employed together with the wall functions implemented. In order to resolve the abrupt fluctuations experienced by the turbulent energy and other parameters located at near wall regions and shear layers a particularly fine computational mesh is necessary which inevitably increases the computer storage and run-time requirements. The PHYSICA Finite Volume code was used for this investigation. With the exception of the k-ε and k-ω models which are available as standard within PHYSICA, all other turbulence models mentioned were implemented via the source code by the authors. The LVEL, LVEL CAP, Wolfshtein, k-ε, k-ω, SST and kε/kl models are described and compared with experimental data.

Airborne transport of Saharan dust to the Mediterranean and to the Atlantic

The Sahara desert is a significant source of particulate pollution not only to the Mediterranean region, but also to the Atlantic and beyond. In this paper, PM 10 exceedences recorded in the UK and the island of Crete are studied and their source investigated, using Lagrangian Particle Dispersion (LPD) methods. Forward and inverse simulations identify Saharan dust storms as the primary source of these episodes. The methodology used allows comparison between this primary source and other possible candidates, for example large forest fires or volcanic eruptions. Two LPD models are used in the simulations, namely the open source code FLEXPART and the proprietary code HYSPLIT. Driven by the same meteorological fields (the ECMWF MARS archive and the PSU/NCAR Mesoscale model, known as MM5) the codes produce similar, but not identical predictions. This inter-model comparison enables a critical assessment of the physical modelling assumptions employed in each code, plus the influence of boundary conditions and solution grid density. The outputs, in the form of particle concentrations evolving in time, are compared against satellite images and receptor data from multiple ground-based sites. Quantitative comparisons are good, especially in predicting the time of arrival of the dust plume in a particular location.

Numerical investigation of a source extraction technique based on an acoustic correction method

An aerodynamic sound source extraction from a general flow field is applied to a number of model problems and to a problem of engineering interest. The extraction technique is based on a variable decomposition, which results to an acoustic correction method, of each of the flow variables into a dominant flow component and a perturbation component. The dominant flow component is obtained with a general-purpose Computational Fluid Dynamics (CFD) code which uses a cell-centred finite volume method to solve the Reynolds-averaged Navier–Stokes equations. The perturbations are calculated from a set of acoustic perturbation equations with source terms extracted from unsteady CFD solutions at each time step via the use of a staggered dispersion-relation-preserving (DRP) finite-difference scheme. Numerical experiments include (1) propagation of a 1-D acoustic pulse without mean flow, (2) propagation of a 2-D acoustic pulse with/without mean flow, (3) reflection of an acoustic pulse from a flat plate with mean flow, and (4) flow-induced noise generated by the an unsteady laminar flow past a 2-D cavity. The computational results demonstrate the accuracy for model problems and illustrate the feasibility for more complex aeroacoustic problems of the source extraction technique.