Martin Buchholz

ls1 mardyn: The Massively Parallel Molecular Dynamics Code for Large Systems

The molecular dynamics simulation code ls1 mardyn is presented. It is a highly scalable code, optimized for massively parallel execution on supercomputing architectures and currently holds the world record for the largest molecular simulation with over four trillion particles. It enables the application of pair potentials to length and time scales that were previously out of scope for molecular dynamics simulation. With an efficient dynamic load balancing scheme, it delivers high scalability even for challenging heterogeneous configurations. Presently, multicenter rigid potential models based on Lennard-Jones sites, point charges, and higher-order polarities are supported. Due to its modular design, ls1 mardyn can be extended to new physical models, methods, and algorithms, allowing future users to tailor it to suit their respective needs. Possible applications include scenarios with complex geometries, such as fluids at interfaces, as well as nonequilibrium molecular dynamics simulation of heat and mass transfer.

Software design for a highly parallel molecular dynamics simulation framework in chemical engineering

Highlights ► Software structure of a nanofluidics simulation program in chemical engineering. ► Design supports especially testing and comparison of different methods. ► Focus on modules for parallelisation using MPI and allowing hybrid parallelisation. ► Supports different libraries (e.g. TBB, OpenMP) for memory-coupled parallelisation. Abstract The software structure of MarDyn, a molecular dynamics simulation program for nanofluidics in chemical engineering, is presented. Multi-component mixtures in heterogeneous states with huge numbers of particles put great challenges on the simulation of scenarios in this field, which cannot be tackled with the established molecular simulation programs. The need to develop a new software for such simulations with an interdisciplinary team opened the chance of using state-of-the-art methods on the modelling as well as on the simulation side. This entails the need to test and compare different methods in all parts of the program to be able to find the best method for each task. It is shown how the software design of MarDyn supports testing and comparing of various methods in all parts of the program. The focus lies on those parts concerning parallelisation, which is on the one hand a pure MPI parallelisation and on the other hand a hybrid approach using MPI in combination with a memory-coupled parallelisation. For the latter, MarDyn not only allows the use of different algorithms, but also supports the use of different libraries such as OpenMP and TBB.

Grid-Supported Simulation of Vapour-Liquid Equilibria with GridSFEA

In order to benefit from grid computing, software applications in CSE often need to be substantially modified or rewritten to a large extent. To reduce the required grid know-how and effort the computational scientist (end user and software developer) needs for this task, we developed a framework for engineering simulations in grid environments (GridSFEA). This paper presents two novel features of GridSFEA: the integrated support for parameter investigations and the controlled execution of long-running simulations in grids. They allow the grid enabling of CSE applications with minimal or even without changes of their source code. Furthermore, the overhead for working in grid environments introduced by our approach, compared to working on classical HPC platforms, is very low. We provide two examples of using GridSFEA for performing vapour-liquid equilibria (VLE) simulations using Molecular Dynamics and Monte Carlo methods. To develop VLE models, parameter investigations are carried out. Large VLE scenarios are computed over a long time, to create test cases for the development of HPC software.

Application of Machine Learning Techniques to the Re-ranking of Search Results

Even though search engines cover billions of pages and perform quite well, it is still difficult to find the right information from the returned results. In this paper we present a system that allows a user to re-rank the results locally by augmenting a query with positive example pages. Since it is not always easy to come up with many example pages, our system aims to work with only a couple of positive training examples and without any negative ones. Our approach creates artificial (virtual) negative examples based upon the returned pages and the example pages before the training commences. The list of results is then re-ordered according to the outcome from the machine learner. We have further shown that our system performs sufficiently well even if the example pages belong to a slightly different (but related) domain.

Group Decision Making

This chapter will deal with situations involving several possibilities that can and should be ranked—for example parties at an election, singers at a singing competition or the variations given in the traffic routing in a city.

Stochastic Traffic Simulation

The third variant for the simulation of traffic considers particular characteristics of traffic such as those observable in computing systems.

Macroscopic Simulation of Road Traffic

Road traffic concerns all of us since everyone is affected by it. However, our wishes and ideas with respect to road traffic are typically somewhat contradictory: On the one hand, we want as much of it as possible. We want to be mobile and get as quickly and comfortably from the apartment to the university or to work, from the hometown to the vacation place or from shopping to the leisure activity. For this, we desire well-developed roads and parking places and convenient and immediately available public transportation. On the other hand, we also want as little of it as possible. Road traffic should not bother us. We do not want to be in commuter traffic nor in the kilometer long traffic jams at the beginning and ending of holidays. We wish to be without the noise and exhaust emissions and prefer green corridors over asphalt streets.

Required Tools in Short

In this chapter, we will give an itemized overview of the topics in mathematics and computer science which we will reference repeatedly in subsequent parts of this book. The selection of topics for this chapter is guided by the simple rule that everything occurring multiple times is introduced only once in this central location, whereas tools that are required only for individual application scenarios will be covered in that particular part. Since our book is primarily directed toward students studying computer science, engineering, or natural sciences who are upper division bachelor students, we assume that most topics of this chapter have been studied in their respective introductory courses in mathematics. On the other hand, the experience with courses on modeling and simulation has shown that a certain “warm-up” period in order to better adjust preliminary knowledge can do no harm—especially in view of the wide mix in academic backgrounds oftentimes observed (and intended so) in our audience.

Global Illumination in Computer Graphics

In this chapter, we turn to a primordial computer science application of modeling and simulation—yet simultaneously we deal with physics on the computer. After all, one of the biggest goals of computer graphics is photo realism, i.e., the generation of computer pictures that appear as realistic as possible. We thereby will encounter models in many places—the description of objects and effects—as well as simulations—with their efficient graphical representation.

Makroskopische Simulation von Straßenverkehr

Wie in Kap. 7 wollen wir erneut Strasenverkehr modellieren und simulieren. Naturlich haben wir wieder das Ziel, Strasenverkehr besser zu verstehen, und unser Modell soll moglichst gut Verkehrsphanomene erklaren und realen Verkehr simulieren konnen. Das Modell soll es uns ermoglichen, Anforderungen an den Verkehr zu optimieren (regelnd einzugreifen) und Veranderungen (zum Beispiel Baumasnahmen) zu planen – ohne alle denkbaren Varianten in der Realitat ausprobieren zu mussen. Dass die verschiedenen Anforderungen zum Teil miteinander konkurrieren, versteht sich dabei fast von selbst. „Freie Fahrt bei leeren Strasen“ aus Sicht des Verkehrsteilnehmers vertragt sich beispielsweise nicht mit dem Ziel des Verkehrsplaners, moglichst vielen Fahrzeugen pro Zeit ohne Stau die Benutzung eines Autobahnabschnitts zu ermoglichen.