TASK Quarterly   Scientific Bulletin of the Centre of Informatics - Tricity Academic Supercomputer & networK   ISSN 1428-6394

Volume 22, Number 4, 2018


Contents:



Abstracts:

  • Dominik Pielak, Mateusz Kowalski and Jacek Lebiedź 3D Model Preparing Patterns for Interactive Urban Visualization

    While working on architectural visualizations, the software developer often has to work with graphic designers who create models in a different environment what can cause many complications. For this reason, it is very important to have some guidelines which can protect both the developer and the designer from commixing mistakes. The paper presents a list of such guidelines based on the authors' experience. The reader can treat the paper as a first step in the development of a service based on cloud computing that verifies the correctness of graphical data in urban visualizations.

  • Henryk Krawczyk and Narek Parsamyan Mashup Approaches for Improving Service Oriented Business Scenarios

    Two kinds of business scenarios are considered: application- and service-oriented. It is shown that the former are more flexible and easier reconfigurable for both traditional (human-oriented) and IT services. Moreover, complex IT services can be created by simple IT services using the modern mashup technology. It is the reason why such business scenarios can be modeled by BPMN in much easier ways. A simple example of a supply chain of a sales company is considered and the available advantages of such technologies are emphasized in the paper.

  • Krzysztof Drypczewski, Jerzy Proficz and Andrzej Stepnowski Influence of YARN Schedulers on Power Consumption and Processing Time for Various Big Data Benchmarks

    Climate change caused by human activities can influence the lives of everybody on the planet. The environmental concerns must be taken into consideration by all fields of study including ICT. Green Computing aims to reduce negative effects of IT on the environment while, at the same time, maintaining all of the possible benefits it provides. Several Big Data platforms like Apache Spark or YARN have become widely used in analytics and High-Performance Computing systems due to the reliability and usability of Map Reduce implementations. The authors research the power consumption and energy efficiency of Hadoop YARN schedulers using Apache Spark under three different workloads. The test cases include: sorting large binary files, counting unique words in large text files and processing satellite imagery from the Sentinel-2 mission. The presented results show small (2%–11%) but distinct differences in the power consumption of FIFO and FAIR schedulers.

  • Piotr Orzechowski TASK Cloud infrastructure in the Centre of Informatics – Tricity Academic Supercomputer \& Network

    The cloud solution called the TASK cloud is described. Its main components and the way of their implementation are described. Examples of deployed environments based on the cloud are also presented. Moreover, the idea of combining the cloud and big data platforms is suggested and discussed.

  • Łukasz Wiszniewski Suitability of LoRaWAN Technology for the Development of Maritime Applications

    The LoRaWAN Technology opens new possibilities for gathering and analysis of distributed data. In the paper we concentrate on its maritime usability which was tested by us in the period from June to August 2018. Measurements of the LoRaWAN network coverage in the Bay of Gdansk area were carried out. Various conditions and places were tested. The research was planned in such a way as to gradually increase the range and control the impact of environmental factors. The results of the measurements confirm the wide application and a broad range of the technology. Combined with cloud computing it leads to new kinds of services and applications.

  • Mirosław Kupczyk, Michał Białoskórski, Rafał Tylman, Klemens Noga, Agnieszka Kwiecień, Mariusz Uchroński and Norbert Meyer PRACE Research Infrastructure Offer for Polish R\&D Community

    The mission of the PRACE Research Infrastructure is to enable high impact scientific discovery and engineering research and development across all disciplines to enhance European competitiveness for the benefit of society. The paper presents the current state of the computational infrastructure, which is unique on the European level and procedures helping with the seamless access to the European HPC infrastructure: supercomputers, applications, storage. Several Polish computational grants have been called in the paper in order to present examples of the Polish domestic scientific engagement.

  • Adam Liwo LARGE-SCALE SIMULATIONS IN CHEMISTRY AND BIOCHEMISTRY

    Computer simulations are nowadays critical in science, medicine, and economy, enabling us to model the phenomena that are too complex to be comprehended based on the equations and otherwise formulated scientific principles given the proverbial “own brains, a piece of paper, and a pencil”. Owing to the development of supercomputers, quantum chemistry could reach beyond single atoms and small molecules to study the chemical reactions that occur in living organisms and in the environment or that are important in industry. Still larger systems can be simulated with the use of empirical force fields at various levels of resolution, from atomistically detailed to aggressive coarse-graining in which groups of atoms are merged into single sites. Owing to computer simulations, new drugs and therapies or materials with desired properties can be designed without having to spend chemicals, money and sacrificing thousands of hours of labor. This article outlines the physical principles behind computer modeling and selected applications of computer simulations in chemistry and biochemistry.

  • Jaromir Jakacki, Anna Przyborska and Mirosław Darecki Supporting Marine Environment Research by Modeling

    The Centre of Informatics – Tri-City Academic Supercomputer and Network (CI TASK) was founded over twenty-five years ago, in 1994. It was a very important initiative undertaken as a result of rapid development of computer technology and one of its main goals was to integrate the computer resources of scientific communities in the Tri-City. CI TASK started to provide computing resources and the first steps in the world of high-performance computing must have been proved to be a significant challenge. The scientific community of the Institute of Oceanology has been cooperating with that computing centre since the very beginning. Currently, CI TASK provides the widest set of tools and hardware that can be used for ocean modeling purposes. The main goal of this work is expression of appreciation and gratitude for the people who have contributed to the developing of this institution and helped in the work using their tools. CI TASK has been helpful in a great number of past and present tasks, however, in this paper we would like to present only three results of our work – a coupled ice-ocean model of the Baltic Sea, assessment of contamination for potential leakage from chemical munition dumped after the Second World War, and the currently operating SatBaltic system.