DSC: Scheduling Parallel Tasks on an Unbounded Number of Processors
IEEE Transactions on Parallel and Distributed Systems
A New Heuristic for Scheduling Parallel Programs on Multiprocessor
PACT '98 Proceedings of the 1998 International Conference on Parallel Architectures and Compilation Techniques
Triplet: A Clustering Scheduling Algorithm for Heterogeneous Systems
ICPPW '01 Proceedings of the 2001 International Conference on Parallel Processing Workshops
EbitSim: An Enhanced BitTorrent Simulation Using OMNeT++ 4
MASCOTS '11 Proceedings of the 2011 IEEE 19th Annual International Symposium on Modelling, Analysis, and Simulation of Computer and Telecommunication Systems
| Hi-index | 0.00 |
Mobile consumers increasingly expect that their applications be context-aware by sensing their dynamic environment to deliver personalized contents. Context sources include user profiles, web history, and the environment captured by mobile device sensors. The benefits of context-aware applications and services include improved user experiences, higher quality of service, optimized system resource utilization, and smarter recommendations. One application domain that has yet to exploit personal context fully is video streaming. Video services, such as Youtube and Netflix, stream the same video clip to millions of users. However, MOOC (Massive Open Online Courses) providers, such as Udacity, Coursera or edX, offer video interaction capabilities to personalize the viewing experience. In this paper, we present the design and evaluation of NewsCast, an application to provide personalized interactive video streaming in real-time. We show how to design, adapt, and customize the NewsCast streaming experience using the Gstreamer open source software. To evaluate the performance of our context-aware video streaming platform, we subjected NewsCast to several deployment scenarios to compare key indicators such as latency, bandwidth and processor utilization. This NewsCast project grew out of the NSERC Strategic Network for Smart Applications on Virtual Infrastructure (SAVI) and our quest to build canonical, data-intensive future Internet applications.