Browsing by Author "Deek, Lara B."

Now showing 1 - 2 of 2

Exploiting Locality of Interest in Online Social Networks
(ACM CoNEXT, 2010) Wittie, Mike P.; Pejovic, Veljko; Deek, Lara B.; Almeroth, Kevin C.; Zhao, Ben Y.
Online Social Networks (OSN) are fun, popular, and socially significant. An integral part of their success is the immense size of their global user base. To provide a consistent service to all users, Facebook, the world’s largest OSN, is heavily dependent on centralized U.S. data centers, which renders service outside of the U.S. sluggish and wasteful of Internet bandwidth. In this paper, we investigate the detailed causes of these two problems and identify mitigation opportunities. Because details of Facebook’s service remain proprietary, we treat the OSN as a black box and reverse engineer its operation from publicly available traces. We find that contrary to current wisdom, OSN state is amenable to partitioning and that its fine grained distribution and processing can significantly improve performance without loss in service consistency. Through simulations of reconstructed Facebook traffic over measured Internet paths, we show that user requests can be processed 79% faster and use 91% less bandwidth. We conclude that the partitioning of OSN state is an attractive scaling strategy for Facebook and other OSN services.
Exploiting Parallel Networks Using Dynamic Channel Scheduling
(Wireless Internet Conference (WICON), 2008-11) Deek, Lara B.; Almeroth, Kevin C.; Wittie, Mike P.; Harras, Khaled A.
Many researchers have been focusing on the outcomes and consequences of the rapid increase and proliferation of mobile wireless technologies. If it is not already the case, it will soon be rare for a user to be in a situation where absolutely no network connection exists. In fact, through numerous devices, users will soon expect to be connected in all places at all times. Through the great variety and increase in the capabilities of these devices, it is not a surprise to find a single user with many connection opportunities. As a result, we believe that the next major evolution of wireless mobile networks will be in the exploitation of multiple network connections in parallel. Due to network heterogeneity, the major challenge in such situations is to determine the way that these networks can be utilized to better serve different network applications. In this work, we propose a dynamic channel scheduling mechanism that adapts to the state of the available channels to provide more efficient usage of network connectivity. We do so by observing channel throughput, creating a set of channel usage combinations, and then choosing the most efficient combination. We evaluate an implementation of the proposed mechanism using emulation. Our results show that under realistic conditions our dynamic approach greatly improves cost delay metrics, and the overall user-perceived performance compared to a more static approach.