August 10, 2016 16:48
As part of the work for project B3 the survey on Opportunities and Challenges for Distributed Data Analysis has now been published by Marco Stolpe at ACM SIGKDD.
This survey motivates how the real-time analysis of data, embedded into the Internet of Things (IoT), enables entirely new kinds of sustainable applications in sectors such as manufacturing, transportation and distribution, energy and utilities, the public sector as well as in healthcare. It presents and discusses the challenges of real-time constraints for state-of-the-art analysis methods. Current research strongly focuses on cloud-based big data analysis. Our survey provides a more balanced view, taking also into account highly communication-constrained scenarios which require research on decentralized analysis algorithms. These must analyse data directly on sensors and small devices. Discussed is the vertical partitioning of data common for the IoT, which is particularly challenging, since information about observations is assessed at different networked nodes. The paper includes a comprehensive bibliography that should provide readers with a good starting point for their own work.
