On analyzing quality of data influences on performance of finite elements driven computational simulations

Authors:
Michael Reiter;Hong-Linh Truong;Schahram Dustdar;Dimka Karastoyanova;Robert Krause;Frank Leymann;Dieter Pahr
Affiliations:
Institute of Architecture of Application Systems, Universität Stuttgart, Germany;Distributed Systems Group, Vienna University of Technology, Austria;Distributed Systems Group, Vienna University of Technology, Austria;Institute of Architecture of Application Systems, Universität Stuttgart, Germany;Institute of Applied Mechanics (CE), Universität Stuttgart, Germany;Institute of Architecture of Application Systems, Universität Stuttgart, Germany;Institute of Lightweight Design and Structural Biomechanics, Vienna University of Technology, Austria
Venue:
Euro-Par'12 Proceedings of the 18th international conference on Parallel Processing
Year:
2012

Citing 5
Cited 0

Efficient parallel computation of unstructured finite element reacting flow solutions

Parallel Computing - Special issue on applications: parallel computing methods in applied fluid mechanics
High performance sparse static solver in finite element analyses with loop-unrolling

Advances in Engineering Software
Data Quality: Concepts, Methodologies and Techniques (Data-Centric Systems and Applications)

Data Quality: Concepts, Methodologies and Techniques (Data-Centric Systems and Applications)
A generic grid interface for parallel and adaptive scientific computing. Part I: abstract framework

Computing
A Novel Framework for Monitoring and Analyzing Quality of Data in Simulation Workflows

ESCIENCE '11 Proceedings of the 2011 IEEE Seventh International Conference on eScience

Quantified Score

Hi-index	0.00

Visualization

Abstract

For multi-scale simulations, the quality of the input data as well as the quality of algorithms and computing environments will strongly impact the intermediate results, the final outcome, and the performance of the simulation. To date, little attention has been paid on understanding the impact of quality of data (QoD) on such multi-scale simulations. In this paper, we present a critical analysis of how QoD influences the results and performance of basic simulation building blocks for multi-scale simulations. We analyze the impact of QoD for Finite Element Method (FEM) based simulation building blocks, and study the dependencies between the QoD of input data and results as well as the performance of the simulation. We devise and implement novel QoD metrics for data intensive, FEM-based simulations and show experiments with real-world applications by demonstrating how QoD metrics can be efficiently used to control and tune the execution of FEM-based simulation at runtime.