Title of monograph: Advances in Dependability Engineering of Complex Systems

Title of chapter: Testing the Significance of Parameters of Models Estimating Execution Time of Parallel Program loops According to the Open MPI Standard

Author/Authors: Łukasz Nozdrzykowski, Magdalena Nozdrzykowska

Year: 2018

Keywords: program loops, time stimation, significance analysis, Rough set theory

Abstract: The authors present their own models for estimating the execution time of program loops in parallel systems with message transmission (Open MPI) along with an analysis of the significance of the parameters used for the estimation and for proving the correctness of these models, which is the main goal of this article. The significance analysis proposed by the authors of the models was based on measurements of actual performance of test loops from the NAS Parallel Benchmarks, which provides sets of test loops in the field of numerical problem solving. By applying the significance analysis, we demonstrate the accuracy of selected parameters as relevant parameters of the presented models. The significance analysis was based on soft reduction of conditional attributes using the relative probability of rules useful in the theory of rough sets.

ISBN: 978-3-319-59414-9

Title: Models for estimating the time of program loop execution in parallel on a CPU and with the use of OpenCL computation on a GPU

Author/Authors: Łukasz Nozdrzykowski, Magdalena Nozdrzykowska

Place of publication: Autobusy. Technika, Eksploatacja, Systemy Transportowe, vol. 226, no. 12, pp. 802-807

Year: 2018

Keywords: pętle programowe, szacowanie czasu wykonania pętli, programowanie CPU i GPGPU, programming loop, estimating the time of the loop, programming CPU and GPGPU

Abstract: The authors present models for estimating the time of execution of program loops compliant with the FAN model with no data dependencies or with data dependencies only within the body programming loop, which can be executed either by CPUs or by stream multiprocessors referred to as GPU cores. The models presented will make it possible to determine whether it would be more efficient to execute computation in the existing environment using the CPU (Central Processing Unit) or a state-of-the-art graphics card with a high-performance GPU (Graphics Processing Unit) and super-fast memory, often implemented in modern graphics cards. Validity checks confirming the developed time estimation model for GPU are presented. The purpose of these models is to provide methods for accelerating the performance of applications performing various tasks, including transport tasks, such as accelerated solution searching, searching paths in graphs, or accelerating image processing algorithms in vision systems of autonomous and semiautonomous vehicles, where these models allow to build an automatic task distribution system between the CPU and the GPU with the variability of computing resources.

Title: Implementation of the attribute significance analysis with the use of soft reduction of attributes in the rough set theory on the basis of the SQL mechanisms

Author/Authors: Łukasz Nozdrzykowski, Magdalena Nozdrzykowska

Place of publication: Measurement Automation Monitoring - Pomiary Automatyka Kontrola, vol. 63, no. 1, pp. 10-14

Year: 2017

Keywords: rough set theory, data analysis, soft reduction of attributes, SQL implementation

Abstract: This article presents a way to use databases supporting the SQL and PL/SQL in the implementation of a method of attribute significance analysis with the use of soft reduction of attributes in the rough set theory. A number of SQL queries are presented, which facilitate the implementation. The original mechanisms presented previously [1] are supplemented with queries which facilitate the execution of attribute coding. The authors present a complete implementation of the method, from the coding of attributes to the determination of the significance of conditional attributes. Application of queries to the database eliminates the necessity to build data grouping and data mining mechanisms and calculation of repetitions of identical rules in the reduced decision rule space. Without the support of a database, the creation of universal data grouping and data mining mechanisms which could be used with any number of attributes is a challenging task