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

Abstract

Abstract-In recent years scientific workflows have been used for conducting data-intensive and long running simulations. Such simulation workflows have processed and produced different types of data whose quality has a strong influence on the final outcome of simulations. Therefore being able to monitor and analyze quality of this data during workflow execution is of paramount importance, as detection of quality problems will enable us to control the execution of simulations efficiently. Unfortunately, existing scientific workflow execution systems do not support the monitoring and analysis of quality of data for multi-scale or multi-domain simulations. In this paper, we examine how quality of data can be comprehensively measured within workflows and how the measured quality can be used to control and adapt running workflows. We present a quality of data measurement process and describe a quality of data monitoring and analysis framework that integrates this measurement process into a workflow management system.

Data Market Design: A Systematic Literature Review

“…This facet entails properties such as featurerichness (i.e., how many dimensions are in the data) [101] and truthfulness (i.e., the extent to which the data corresponds with reality) [118]. Finally, data quality has a contextbased facet [119], which includes the ease with which it can be integrated with other data [34], the relevance of the data for the data consumer, as well as (proof of) data provenance (i.e., the origin of the data) [10], [13].…”

Principles of Software-Defined Elastic Systems for Big Data Analytics

“…However, an important issue is how to scale the problem solving when the complex software detects some critical situations? We approach this question by using human-based workflows [13] and using high-level elasticity control language [11] to invoke humanbased services, when needed, e.g., when the quality of data is low. The following list shows an example of how to invoke human-based services within an analytics service: In this case, both software and humans are involved and human-based services (indicated by ServiceUnitType.HBS) are scaled out to examine the situations.…”

On Analyzing Quality of Data Influences on Performance of Finite Elements Driven Computational Simulations

“…In order to measure, monitor, and evaluate QoD metrics for FEM-based simulations, we utilize our extensible QoD Evaluation Framework for workflows developed in [8]. Generally, with this framework, we can determine QoD in a very flexible way: (i) platform and language independent metrics and interpretations can be invoked, (ii) separate metrics as well as metrics that are included in comprehensive algorithms (e.g.…”

“…We consider QoD as a tuple of characteristics and goodness [8]. A characteristics of data will be analyzed without any simulation context, while the goodness of the characteristics of data will be evaluated with respect to the specific simulation context.…”

“…Some of them, e.g. Ansys 1 , use pre-implemented functions and have a strong FEM focus Time Loop (8) for i = 1 to n A FEM-based multi-scale simulation can have millions of basic simulation building blocks, e.g. a human skeleton has approximately 206 bones, each bone has thousands to millions of cells, where each cell can be simulated by a FEMbased simulation building block.…”