Brent Gorda

Writing Scientific Software &#x2013; A Guide to Good Style, by Suely Oliveira and David Stewart

Grid brings the power of many computers to scientists. However, the development of Grid-enabled applications requires knowledge about Grid infrastructure and low-level API to Grid services. In turn, workflow management systems provide a high-level environment for rapid prototyping of experimental computing systems. Coupling Grid and workflow paradigms is important for the scientific community: it makes the power of the Grid easily available to the end user. The paradigm of data driven workflow execution is one of the ways to enable distributed workflow on the Grid. The work presented in this paper is carried out in the context of the Virtual Laboratory for e-Science project. We present the VLAM-G workflow management system and its core component: the Run-Time System (RTS). The RTS is a dataflow driven workflow engine which utilizes Grid resources, hiding the complexity of the Grid from a scientist. Special attention is paid to the concept of dataflow and direct data streaming between distributed workflow components. We present the architecture and components of the RTS, describe the features of VLAM-G workflow execution, and evaluate the system by performance measurements and a real life use case.

Vladimir Korkhov

Dmitry Vasyunin

Adianto Wibisono

Adam S.Z. Belloum

M&#xe1;rcia A. Inda

Marco Roos

Timo M. Breit

L.O. Hertzberger

VLAM-G: Interactive Data Driven Workflow Engine for Grid-Enabled Resources

A vectorized Matlab implementation for the linear finite element is provided for the two-dimensional linear elasticity with mixed boundary conditions. Vectorization means that there is no loop over triangles. Numerical experiments show that our implementation is more efficient than the standard implementation with a loop over all triangles.

Jonas Koko

Vectorized Matlab Codes for Linear Two-Dimensional Elasticity

On many parallel target platforms it can be advantageous to implement parallel applications as a collection of multiprocessor tasks that are concurrently executed and are internally implemented with fine-grain SPMD parallelism. A class of applications which can benefit from this programming style are methods for solving systems of ordinary differential equations. Many recent solvers have been designed with an additional potential of method parallelism, but the actual effectiveness of mixed task and data parallelism depends on the specific communication and computation requirements imposed by the equation to be solved. In this paper we study mixed task and data parallel implementations for general linear methods realized using a library for multiprocessor task programming. Experiments on a number of different platforms show good efficiency results.

Thomas Rauber

Gudula R&#xfc;nger

Mixed Task and Data Parallel Executions in General Linear Methods

Causal modeling and the accompanying learning algorithms provide useful extensions for in-depth statistical investigation and automation of performance modeling. We enlarged the scope of existing causal structure learning algorithms by using the form-free information-theoretic concept of mutual information and by introducing the complexity criterion for selecting direct relations among equivalent relations. The underlying probability distribution of experimental data is estimated by kernel density estimation. We then reported on the benefits of a dependency analysis and the decompositional capacities of causal models. Useful qualitative models, providing insight into the role of every performance factor, were inferred from experimental data. This paper reports on the results for a LU decomposition algorithm and on the study of the parameter sensitivity of the Kakadu implementation of the JPEG-2000 standard. Next, the analysis was used to search for generic performance characteristics of the applications.

Jan Lemeire

Erik Dirkx

Frederik Verbist

Causal Analysis for Performance Modeling of Computer Programs

Robert C. Kirby

Understanding and Implementing the Finite Element Method, by Mark Gockenbach

Up to 1920 processors of a cluster of distributed shared memory machines at the NASA Ames Research Center are being used to simulate ocean circulation globally at horizontal resolutions of 1/4, 1/8, and 1/16-degree with the Massachusetts Institute of Technology General Circulation Model, a finite volume code that can scale to large numbers of processors. The study aims to understand physical processes responsible for skill improvements as resolution is increased and to gain insight into what resolution is sufficient for particular purposes. This paper focuses on the computational aspects of reaching the technical objective of efficiently performing these global eddy-resolving ocean simulations. At 1/16-degree resolution the model grid contains 1.2 billion cells. At this resolution it is possible to simulate approximately one month of ocean dynamics in about 17 hours of wallclock time with a model timestep of two minutes on a cluster of four 512-way NUMA Altix systems. The Altix systems' large main memory and I/O subsystems allow computation and disk storage of rich sets of diagnostics during each integration, supporting the scientific objective to develop a better understanding of global ocean circulation model solution convergence as model resolution is increased.

Chris Hill

Dimitris Menemenlis

Bob Ciotti

Chris Henze

Investigating Solution Convergence in a Global Ocean Model Using a 2048-Processor Cluster of Distributed Shared Memory Machines

Autonomous rover navigation is a critical technology for robotic exploration of Mars. Simulation allows more extensive testing of such technologies than would be possible with hardware test beds alone. A large number of simulations, running in parallel, can test an algorithm under many different operating conditions to quickly identify the operational envelope of the technology and identify failure modes that were not discovered in more limited testing. GESTALT is the autonomous navigation algorithm developed for NASA's Mars rovers. ROAMS is a rover simulator developed to support the Mars program. We have integrated GESTALT into ROAMS to test closed-loop, autonomous navigation in simulation. We have developed a prototype capability to run many copies of ROAMS in parallel on a supercomputer, varying input parameters to rapidly explore GESTALT's performance across a parameter space. Using these tools, we have demonstrated that large scale simulation can identify performance limits and unexpected behaviors in an algorithm. Such parallel simulation was able to test approximately 500 parameter combinations in the time required for a single test on a hardware test bed.

Richard Madison

Abhinandan Jain

Christopher Lim

Mark Maimone

Performance Characterization of a Rover Navigation Algorithm Using Large-Scale Simulation

Boleslaw K. Szymanski

Fortran Programming Language and Scientific Programming: 50 Years of Mutual Growth

About this journal

Scientific Programming provides a forum for research results in, and practical experience with, software engineering environments, tools, languages, and models of computation aimed specifically at supporting scientific and engineering computing.

Journal profile

Meet the editorial board

Chief Editor&nbsp;Professor Tramontana&nbsp;is based at the University of Catania and his research primarily concerns the areas of software engineering and distributed systems.

Editor spotlight

Submit your proposal

Do you think there is an emerging area of research that really needs to be highlighted? Or an existing research area that has been overlooked or would benefit from deeper investigation? Raise the profile of a research area by leading a Special Issue.

Special Issues

Author guidelines

Editorial board

Databases and indexing

Scientific Programming - Table of contents 2007

title

Scientific Programming provides a forum for research results in, and practical experience with, software engineering environments, tools, languages, and models of computation aimed specifically at supporting scientific and engineering computing.

Scientific Programming