The Selection Problem in Distributed Computing Systems with Application to Internet Environments
Dmitry Korzun received B.Sc(1997) and M.Sc(1999) degrees in Applied Mathematics and Computer Science from the Petrozavodsk State University (PetrSU, Russia). He received a Ph.D. degree in Physics and Mathematics from the St.-Petersburg State University (Russia) in 2002. He is an Associate Professor at Department of Computer Science of PetrSU (since 2003). He was Visiting Research Scientist at the Helsinki Institute for Information Technology HIIT, Aalto University, Finland (2005-2014). In 2014-2016 he performed the duties of Vice-dean for Research at Faculty of Mathematics and Information Technology of PetrSU. Since 2014 he has acted as Leading Research Scientist at PetrSU, originating research and development activity within fundamental and applied research projects on emerging topics in ubiquitous computing, smart spaces, and Internet technology. He serves on technical program committees and editorial boards of many international conferences and journals. He is an author of more than 150 research and educational publications.
No video of the event yet, sorry!
Components of a distributed system are subject to autonomous and cooperative activity within the system. To implement this activity, a component applies an individual strategy when some reasonable actions are selected based on the observed system state and other components' activity. In the Internet case, this selection problem widely accepts the multi-agents system approach from Artificial Intelligence (AI) as well the smart object approach from Internet of Things (IoT). In this talk, we consider several application problem instances of the selection problem, which come from our research and development experience of development for Internet environments. We introduce mathematical models related to the studied Internet application problems. We show the important role of ranking models from AI, when available information is used to evaluate some ranks for objects under the selection and the highest-rank objects provide the solution.
The following application problems are introduced. 1. Multipath routing of a continuous data flow, i.e., selection of paths for faster multimedia transfer trough the Internet. 2. Data interleaving in mobile traffic transmission networks, i.e., selection of low-used slots in the channel for improving the network utilization. 3. Notifying on asynchronous updates occurred in the smart space, i.e., selection of check time for active control of the shared information state in IoT environment. 4. Edge-centric task delegation , i.e., selection of appropriate surrounding devices as volunteers to perform processing tasks and to make the IoT environment cooperative.
- 2018 September 22 - 16:20
- 20 min
- Stochastic Modeling and Applied Research of TechnologY
- 1. Stochastic Modeling and Applications