Performance Evaluation of Parallel Haemodynamic Computations on Heterogeneous Clouds

Authors

  • Oleg Bystrov Vilnius Gedinimas Technical University, Vilnius 10223
  • Arnas Kačeniauskas Vilnius Gedinimas Technical University, Vilnius 10223
  • Ruslan Pacevič Vilnius Gedinimas Technical University, Vilnius 10223
  • Vadimas Starikovičius Vilnius Gedinimas Technical University, Vilnius 10223
  • Algirdas Maknickas Vilnius Gedinimas Technical University, Vilnius 10223
  • Eugeniuš Stupak Vilnius Gedinimas Technical University, Vilnius 10223
  • Aleksandr Igumenov Vilnius Gedinimas Technical University, Vilnius 10223

Keywords:

Cloud computing, parallel computing, haemodynamic flows, parallel performance analysis, energy consumption, bi-objective optimization problem

Abstract

The article presents performance evaluation of parallel haemodynamic flow computations on heterogeneous resources of the OpenStack cloud infrastructure. The main focus is on the parallel performance analysis, energy consumption and virtualization overhead of the developed software service based on ANSYS Fluent platform which runs on Docker containers of the private university cloud. The haemodynamic aortic valve flow described by incompressible Navier-Stokes equations is considered as a target application of the hosted cloud infrastructure. The parallel performance of the developed software service is assessed measuring the parallel speedup of computations carried out on virtualized heterogeneous resources. The performance measured on Docker containers is compared with that obtained by using the native hardware. The alternative solution algorithms are explored in terms of the parallel performance and power consumption. The investigation of a trade-off between the computing speed and the consumed energy is performed by using Pareto front analysis and a linear scalarization method.

Downloads

Download data is not yet available.

Author Biographies

Oleg Bystrov, Vilnius Gedinimas Technical University, Vilnius 10223

Oleg Bystrov is a Ph.D. student in Informatics Engineering at Vilnius Gediminas Technical University (VGTU), Lithuania, and System Administrator at the Laboratory of Security of Information Technologies. He received M.Sc. in Informatics Engineering from VGTU in 2010. His research interests include Distributed and Cloud Computing, Green Computing, Performance Evaluation, Virtualization Technologies, IaaS, OpenStack, Docker, LXC, Operating Systems and Networking.

Arnas Kačeniauskas, Vilnius Gedinimas Technical University, Vilnius 10223

Arnas Kačeniauskas is the Director of the Institute of Applied Computer Science of Vilnius Gediminas Technical University (VGTU), Lithuania, and the Chief Researcher at the Laboratory of Parallel Computing. He also is Professor and Ph.D. supervisor at the Department of Graphical Systems at VGTU. He received Professor in Informatics Engineering and Ph.D. in Mechanical Engineering at VGTU. He is R & D Project Manager, author and co-author of 34 scientific papers indexed in Clarivate Analytics WoS database journals. His research interests include Parallel, Distributed, Grid and Cloud Computing, High-Performance Computing, Performance of SaaS, Linux Containers, CFD, Haemodynamics, Coupled Problems in Multi-Physics, GPGPU.

Ruslan Pacevič, Vilnius Gedinimas Technical University, Vilnius 10223

Ruslan Pacevič is the Associated Professor at the Department of Graphical Systems of Vilnius Gediminas Technical University (VGTU), Lithuania. He also is the postdoctoral research fellow at the Department of Applied Informatics of Kaunas University of Technology. He received Ph.D. in Informatics Engineering from VGTU in 2015. He is the co-author of several scientific papers and participant of several European and national research projects. His research interests include Distributed, Grid and Cloud Computing, Development of SaaS and Middleware Components, OpenStack, Eucalyptus, Visualization Software, GPGPU, OpenCL.

Vadimas Starikovičius, Vilnius Gedinimas Technical University, Vilnius 10223

Vadimas Starikovičius received his Ph.D. degree in mathematics from Vilnius University, Lithuania in 2002. Currently, he is the Professor at the Department of Mathematical Modelling and the Head of Laboratory of Parallel Computing at Vilnius Gediminas Technical University, Lithuania. His current research interests include parallel and distributed computing, numerical methods for solution of partial differential equations. He has published over 24 refereed articles in these areas.

Algirdas Maknickas, Vilnius Gedinimas Technical University, Vilnius 10223

Algirdas Maknickas is the Senior Researcher at the Institute of Mechanics and the Head of the Laboratory of Numerical Simulations of Vilnius Gediminas Technical University (VGTU), Lithuania. He also is Professor and Ph.D. supervisor at the Departments of Biomechanical Engineering and Mechanical and Material Engineering at VGTU. He received Ph.D. in Mechanical Engineering at VGTU. He is author and co-author of 23 articles and 16 proceeding papers indexed in Clarivate Analytics WoS database journals. His research interests include Biomechanical Engineering, Haemodynamics, Geometric Modelling, Applications of High-Performance Computing, Linear and non-linear Continuum Mechanics, Coupled Problems in Multiphysics, Artificial Intelligence and Computational Complexity, GPGPU.

Eugeniuš Stupak, Vilnius Gedinimas Technical University, Vilnius 10223

Eugeniuš Stupak is the Associated Professor at the Department of Applied Mechanics of Vilnius Gediminas Technical University (VGTU), Lithuania. He received Ph.D. in Mechanical Engineering from VGTU in 2004. He is the co-author of several scientific papers and participant of several national research projects. His research interests include Advanced Mesh Generation Strategies, Patient-Specific Modelling, Solution of Coupled Multiphysics Problems.

Aleksandr Igumenov, Vilnius Gedinimas Technical University, Vilnius 10223

Aleksandr Igumenov is a lecturer in the Department of Information Technologies, at Vilnius Gediminas Technical University, Lithuania. He received Ph.D. in Informatics Engineering from the Vilnius University in 2012. He is the author and co-author of several scientific papers. His main research interests include Green and Energy Efficient Computing, High-Performance Computing, IoT and Internet Technologies, Blockchain Technologies, Global Optimization.

Downloads

Published

2021-01-12

How to Cite

Bystrov, O., Kačeniauskas, A., Pacevič, R., Starikovičius, V., Maknickas, A., Stupak, E., & Igumenov, A. (2021). Performance Evaluation of Parallel Haemodynamic Computations on Heterogeneous Clouds. COMPUTING AND INFORMATICS, 39(4), 695–723. Retrieved from http://www.cai.sk/ojs/index.php/cai/article/view/2020_4_695

Issue

Section

Special Section Articles