Scholarly Work - Computer Science
Permanent URI for this collectionhttps://scholarworks.montana.edu/handle/1/3034
Browse
6 results
Search Results
Item AirLab: Distributed Infrastructure for Wireless Measurements(USENIX, 2010) Kone, Vinod; Zheleva, Mariya; Wittie, Mike P.; Zhang, Zengbin; Zhao, Xiaohan; Zhao, Ben Y.; Belding, Elizabeth M.; Zheng, Haitao; Almeroth, Kevin C.The importance of experimental research in the field of wireless networks is well understood. So far researchers have either built their own testbeds or accessed third-party controlled testbeds (http://orbit-lab.org) or used publicly available traces (http://crawdad.cs.dartmouth.edu) for evaluation. While immensely useful, all these approaches have their drawbacks. While building own test beds requires cost and effort, third-party controlled test beds do not replicate real network deployments. On the other hand, the publicly available traces are often collected using different software and hardware platforms, making it very difficult to compare results across traces. As a result, observations are often inconsistent across different networks, leading researchers to draw potentially conflicting conclusions across their own studies. To facilitate meaningful analysis of wireless networks and protocols, we need a way to collect measurement traces across a wide variety of network deployments, all using a consistent set of measurement metrics. Widespread multi-faceted data collection will provide multiple viewpoints of the same network, enabling deeper understanding of both self and exterior interference properties, spectrum usage, network usage, and a wide variety of other factors. Furthermore, data collected in this manner across a variety of heterogeneous network types, such as university, corporate, and home environments, will facilitate cross-comparison of observed network phenomena within each of these settings. To address the critical need for comparable and consistent wireless traces, we propose AirLab, a publicly accessible distributed infrastructure for wireless measurementsItem 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.Item Network Optimization with Dynamic Demands and Link Prices(Allerton Conference, 2012) Patterson, Stacy; Wittie, Mike P.; Almeroth, Kevin C.; Bamieh, BassamWe present Overlapping Cluster Decomposition (OCD), a novel distributed algorithm for network optimization targeted for networks with dynamic demands and link prices. OCD uses a dual decomposition of the global problem into local optimization problems in each node’s neighborhood. The local solutions are then reconciled to find the global optimal solution. While OCD is a descent method and thus may converge slowly in a static network, we show that OCD can more rapidly adapt to changing network conditions than previously proposed first-order and Newton-like network optimization algorithms. Therefore, OCD yields better solutions over time than previously proposed methods at a comparable communication cost.Item IP2DC: Making Sense of Replica Selection Tools(USENIX, 2013) Bharata, Anish; Wittie, Mike P.; Yang, QingCloud-based applications being developed for consumer electronics market (tablets, smart TVs) struggle to deliver the same level of responsiveness as standalone software, leading to user frustration and slow adoption. Often the network that separates user end-hosts from server back-end is to blame. To limit the impact of poor network performance on message delay, or lag, back-end logic and application data are deployed across geographically distributed servers and user requests are directed to the closest one [14]. Such nearby servers deliver content more quickly thanks to a faster expansion of TCP congestion window and more rapid retransmissions over low round-trip time (RTT) paths. Our early results, presented in this poster, show a high level of discrepancy between the available tools and motivate further measurement as well as the need to develop techniques for more accurate server replica selection.Item Cascading Impact of Lag on User Experience in Multiplayer Games(USENIX, 2013) Howard, Eben; Cooper, Clint; Wittie, Mike P.; Yang, QingPlaying cooperative multiplayer games should be fun for everyone involved and part of having fun in games is being able to perform well, be immersed, and stay engaged [13, 17]. These indicators of enjoyment are part of a user's Quality of Experience (QoE), a measure which further includes additional metrics such as attention levels and ability to succeed. Players stop playing the game when it ceases to provide a high enough QoE, especially in cooperative and social games. [8, 18, 19]. Industry application development and current research both operate with the assumption that for any given individual in a group, that individual's QoE is affected only by their own network condition and not the network conditions of the other group members [4, 7, 8]. We show that this assumption is incorrect. Our research shows that the QoE of all group members is negatively affected by a single member's lag (communication delay, or loss caused by poor network conditions). Understanding a user's QoE as a function that includes other users' network conditions has the potential to improve lag mitigation strategies for multiplayer games and other group applications.Item MITATE: Mobile Internet Testbed for Application Traffic Experimentation(Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services (Mobiquitous), 2013) Goel, Utkarsh; Miyyapuram, Ajay; Yang, Qing; Wittie, Mike P.This paper introduces a Mobile Internet Testbed for Application Trac Experimentation (MITATE). MITATE is the first programmable testbed to support the prototyping of application communications between mobiles and cloud datacenters. We describe novel solutions to device security and resource sharing behind MITATE. Finally, we show how MITATE can answer network performance questions crucial to mobile application design.