Taejoong Chung

Bundling practice in BitTorrent: what, how, and why

We conduct comprehensive measurements on the current practice of content bundling to understand the structural patterns of torrents and the participant behaviors of swarms on one of the largest BitTorrent portals: The Pirate Bay. From the datasets of the 120K torrents and 14.8M peers, we investigate what constitutes torrents and how users participate in swarms from the perspective of bundling, across different content categories: Movie, TV, Porn, Music, Application, Game and E-book. In particular, we focus on: (1) how prevalent content bundling is, (2) how and what files are bundled into torrents, (3) what motivates publishers to bundle files, and (4) how peers access the bundled files. We find that over 72% of BitTorrent torrents contain multiple files, which indicates that bundling is widely used for file sharing. We reveal that profit-driven BitTorrent publishers who promote their own web sites for financial gains like advertising tend to prefer to use the bundling. We also observe that most files (94%) in a bundle torrent are selected by users and the bundle torrents are more popular than the single (or non-bundle) ones on average. Overall, there are notable differences in the structural patterns of torrents and swarm characteristics (i) across different content categories and (ii) between single and bundle torrents.

Unveiling group characteristics in online social games: a socio-economic analysis

Understanding the group characteristics in MMORPGs is important in user behavior studies since people tend to gather together and form groups due to their inherent nature. In this paper, we analyze the group activities of users in Aion, one of the largest MMORPGs, based on the records of the activities of 94,497 users. In particular, we focus on (i) how social interactions within a group differ from the ones across groups, (ii) what makes a group rise, sustain, or fall, (iii) how group members join and leave a group, and (iv) what makes a group end. We first find that structural patterns of social interactions within a group are more likely to be close-knit and reciprocative than the ones across groups. We also observe that members in a rising group (i.e., the number of members increases) are more cohesive, and communicate with more evenly within the group than the ones in other groups. Our analysis further reveals that if a group is not cohesive, not actively communicating, or not evenly communicating among members, members of the group tend to leave.

CoRC: coordinated routing and caching for named data networking

Named Data Networking (NDN) uses content names as routing entries, and thus the scalability of NDN routing is of primary concern. NDN allows in-network caching as a built-in functionality; however, if network nodes make caching decisions individually, duplicate copies of the same content may exist among nearby nodes. To address these problems, we propose Coordinated Routing and Caching (CoRC) that mitigates routing scalability and enhances the efficiency of the in-network storage. CoRC aligns the routing and caching mechanisms to manage the same content namespace for better performance. We evaluate CoRC (and its variants) with Vanilla NDN in terms of the cache hit ratio, hop count, and traffic load by running software routers on Amazon EC2. To demonstrate the feasibility of CoRC, we also implement and test the processing time of CoRC forwarding in Linux machines.

How prevalent is content bundling in BitTorrent

Despite the increasing interest in content bundling in BitTorrent systems, there are still few empirical studies on the bundling practice in real BitTorrent communities. In this paper, we conduct comprehensive measurements on one of the largest BitTorrent portals: The Pirate Bay. From the torrents data set collected for 38 days from April to May, 2010, we study how prevalent bundling is and how many files are bundled in a torrent, across different types of contents shared: Movie, Porn, TV, Music, Application, E-book, and Game.

Sharing Topics in Pinterest: Understanding Content Creation and Diffusion Behaviors

Pinterest provides a social curation service where people can collect, organize, and share content (pins in Pinterest) that reflect their interests. This paper investigates (1) the differences in pinning (i.e., the act of posting a pin) and repinning (i.e., the act of sharing other users pin) behaviors by topics and user gender, and (2) the relations among topics in Pinterest. We conduct a measurement study using a large-scale dataset (1.6 M pins shared by 1.1 M users) in Pinterest. We show that there is a notable discrepancy between pinning and repinning behaviors on different topics. We also show that male and female users show different behaviors on different topics in terms of dedication, responsiveness, and sentiment. By introducing the notion of a Topic Network (TN) whose nodes are topics and are linked if they share common users, we analyze how topics are related to one another, which can give a valuable implication on topic demand forecasting or cross-topic advertisement. Lastly, we explore the implications of our findings for predicting a users interests and behavioral patterns in Pinterest.

Systematic Support for Content Bundling in BitTorrent Swarming

Despite the tremendous success of BitTorrent, its swarming system suffers from a fundamental limitation: lower or no availability of unpopular contents. Menasche et al. showed that bundling is one solution to mitigate this availability problem; it improves the availability and reduces download times for unpopular contents by combining multiple files into a single swarm. While current content bundling is done by manually and in an ad-hoc manner decided by publishers, we propose a systematic content bundgling scheme based on content similarity in swarming systems to maximize the gain in availability and download speed. We evaluate three systematic bundling algorithms (cosine, Levenshtein distance, matching coefficient distance) and we find that our systematic bundling scheme increases the number of content files in a swarm, which improves content availability. To the best of our knowledge, this is the first work which raises and explores the possibilities and benefits of systematic content bundling in swarming systems.

Content publishing and downloading practice in bittorrent

BitTorrent has been popular over the last decade. However, few studies have made serious efforts to understand who and why publish torrents, and what strategies are adopted by publishers. In this paper, we study the current content publishing practice in BitTorrent from a socio-economic point of view, by unraveling (1) how files are published by publishers, (2) what strategies are adopted by publishers, and (3) how effective those strategies are. To this end, we conduct comprehensive measurements on one of the largest BitTorrent Portal, The Pirate Bay (TPB). From the datasets of 52 K torrents and 16 M users, we classify the content publishers into three types: (i) fake publishers, (ii) profit-driven publishers, and (iii) altruistic publishers. We show that a significant amount of traffic (61%) of BitTorrent has been generated (i.e., unnecessarily wasted) to download fake torrents. Therefore, we suggest a method to filter out fake publishers on TPB by considering their distinct publishing patterns learned from our measurement study, and show the proposed method can reduce around 45% of the total download traffic. We also reveal that profit-driven publishers adopt different publishing strategies according to their revenue models (e.g., advertising private tracker sites to attract potential new members, or exposing image URLs to make people click the URL links).

Bandwidth Allocation for BitTorrent under Multi-Torrent Environments

BitTorrent has achieved a great success in the field of peer-to-peer (P2P) file sharing. Although BitTorrent allows peers to share files efficiently and scalably, it shows inefficiency when a client participates in multiple torrents where each of them concurrently competing for the limited link bandwidth. In this paper, we propose a new bandwidth allocation algorithm, which greedily increases the bandwidth consumption for downloading, to reduce file transfer time considering the current download/upload status. To compensate overall performance degradation resulting from our greedy allocation, we suggest modifying the choking algorithm of BitTorrent to consider the ratio of seeders and leechers in each torrent. Through comprehensive experiments, we validate the performance gain of the proposed scheme over original BitTorrent in a mix of WiFi and Ethernet testbed and large scale public torrents.

A target-centric surveillance system based on localization and social networking

Surveillance systems are developed to enhance security and safety by constantly observing locations of interest. Although those systems can observe scenes of individual cameras separately, it is difficult to figure out what happened to the target moving across multiple cameras. This paper first proposes Video Diary Service (VDS) to solve this problem. VDS is an automatic video-oriented diary service, which keeps track of users’ lives. In addition, VDS can identify social networking relationships among the users, as well as record videos of the users. By exploiting these properties of VDS, we extend VDS into a new surveillance system called S-VDS. S-VDS is a target-centric surveillance system which focuses on the target, not the area, with its comprehensive information including the location, time, social relationship, and preferences. We then develop the basic functions of the proposed system and demonstrate its feasibility. We also illustrate three applications (i.e., a remote healthcare system, an anti-crime system, and a system for finding missing children), where the proposed system can enhance security and safety by considering individual surveillance purposes.

Strategic bundling for content availability and fast distribution in BitTorrent

Abstract BitTorrent, the immensely successful file swarming system, supports content bundling: a common strategy by which publishers package multiple related files and disseminate them via a single larger swarm. It has been reported that bundling in BitTorrent is wide-spread, currently being done in a subjective and manual manner by individual publishers. This paper is motivated by the following questions: What if bundling is automatically supported by the BitTorrent swarming system (e.g., at a tracker)? By what criteria and how can files be automatically bundled, for better performance of the swarming system? How much performance improvement could be obtained with such an automatic bundling system? To answer the questions, we first suggest nine bundling strategies based on (i) the attributes of the torrents such as title or tags and (ii) time-varying swarm dynamics such as torrent popularity or availability. We then propose and develop a tracker-based bundling system, where all the proposed bundling strategies are implemented and evaluated with a set of real BitTorrent traces. We show that all the proposed bundling strategies outperform no-bundling case in terms of the availability and file download time. The bundling strategy that bundles more popular/available torrents with less popular/available ones outperforms others in most cases. We find that title- and tag-similarity based bundling strategies also provide performance improvement comparable to those of the popularity/availability based bundling strategies.