Big Data Modeling and Analytics for Large-scale Data and Applications

Wednesday, February 11, 2015 - 10:10
TH 425
M. Omair Shafiq, Ph.D. Candidate (University of Calgary)

Monitoring and management of large scale data and applications have always been complex tasks, especially because the data is unstructured and often logged in applications in a syntactic manner. This makes it quite limited, requires manual interpretation and hence makes the process of monitoring and management slow, cumbersome and hard. This talk will present an overview of our proposed solution of data modeling in correlation with analytical techniques in order to improve monitoring and management process for large-scale data and applications. We carry out semantic (i.e., highly structured, formalized and expressive) modeling of data, execution workflow and logs, and then build, customize and use Data Analytics techniques, in correlation with the semantically enriched data, to process the data that helps in automating the monitoring and management process. There have been several related efforts but such approaches still could not achieve the goal effectively mainly because (1) no consideration of correlation between the modeling of data, execution workflow and logs and (2) having data analytical solutions handle unstructured, ambiguous and raw data. We have designed and developed our unique hybrid approach of partially using semantics for data modeling and description, as well as customized data analysis and social network analysis techniques to be able to automatically interpret and process the highly structured information from data and logs generated during the execution. In this way, our approach combines the best characteristics of data modeling and data analytics and helps in improving the automated monitoring and management of data and applications at large-scale. The impact, usefulness and effectiveness of our solution have been demonstrated by applying it on real life use-case scenarios. Our research publications and collaboration with academia and industry have already shown promising results.


M. Omair Shafiq recently defended his PhD in Computer Science, in December 2014, at the University of Calgary. He also completed his Master in Computer Science at the University of Calgary in January 2011. He has over 10 years of extensive experience of working in research, development and teaching at different academic and industrial research institutions in North America, Europe and Asia. His research focuses on Data Modeling (using Semantics and Software Engineering), Big Data Analytics, Social Network Analysis, Web Services and Cloud Computing. He has published over 50 refereed journal, conference, and workshop papers. He has been serving on technical program committees of over 30 conferences and workshops. He has been part of organizing committees of over 8 conference and workshops. He is a recipient of the NSERC Vanier CGS Award (Canada’s most prestigious PhD research award) in 2012, J. B. Hynes Research Innovation Award in 2012, Alberta Innovates – Technology Futures (AITF) Graduate Award in 2010. He also received Department Research Excellence Award in 2009 and Teaching Excellence Award from the Department of Computer Science, University of Calgary in 2010. Before that, he worked as Researcher at Semantic Technology Institute (STI), in Innsbruck Austria (previously known as Digital Enterprise Research Institute) on EU and Austrian government funded ICT research projects. He has proven track record and experience of teaching, research, contributing to successful research funding proposals and grants and maintaining active research collaborations with academia and industry since 2004.