Showing papers by "Carsten Griwodz published in 2013"

Commute path bandwidth traces from 3G networks: analysis and applications

Abstract: In this dataset paper, we present and make available real-world measurements of the throughput that was achieved at the application layer when adaptive HTTP streaming was performed over 3G networks using mobile devices. For the streaming sessions, we used popular commute routes in and around Oslo (Norway) traveling with different types of public transportation (metro, tram, train, bus and ferry). We also have a few logs using a car. Each log provides a times-tamp, GPS coordinates and the measured number of bytes downloaded for approximately every second of the route. The dataset can be used in several ways, but the most obvious application is to emulate the same network bandwidth behavior (on specific geographical positions) for repeated experiments.

Bagadus: an integrated system for arena sports analytics: a soccer case study

Abstract: Sports analytics is a growing area of interest, both from a computer system view to manage the technical challenges and from a sport performance view to aid the development of athletes. In this paper, we present Bagadus, a prototype of a sports analytics application using soccer as a case study. Bagadus integrates a sensor system, a soccer analytics annotations system and a video processing system using a video camera array. A prototype is currently installed at Alfheim Stadium in Norway, and in this paper, we describe how the system can follow and zoom in on particular player(s). Next, the system will playout events from the games using stitched panorama video or camera switching mode and create video summaries based on queries to the sensor system. Furthermore, we evaluate the system from a systems point of view, benchmarking different approaches, algorithms and tradeoffs.

What should you cache?: a global analysis on YouTube related video caching

Abstract: Following advice from the YouTube recommendation system is one of the ways users browse through the videos offered by YouTube. The system presents related videos based on several factors depending on the current video requested. This related videos list can be used by caching infrastructure to reduce network bandwidth consumption. In this paper, we analyze the differences between user-specific recommendation lists. We perform this analysis on 100s of user nodes from all around the world divided into 4 geographical regions using PlanetLab. Based on our analysis, we find that the related videos differ less in the top half (1-10) of the related video list offered by YouTube compared to the bottom half (11-20). Based on our analysis, we suggest that, caching or prefetching of the Top 10 of the related videos is advantageous over a period of time than caching the whole list offered by YouTube.

Cache-centric video recommendation: an approach to improve the efficiency of YouTube caches

Abstract: In this paper, we take advantage of the user behavior of requesting videos from the related list provided by YouTube and the user behavior of requesting videos from the top of this related list to improve the performance of YouTube's caches. We recommend a related list reordering approach which modifies the order of the videos shown on the related list based on the content in the cache. The main goal of our reordering approach is to push the contents already in the cache to the top of the related list and push non-cached contents towards the bottom, which increases the likelihood that the already cached content will be chosen by the viewer. We analyze the benefits of our approach by an investigation that is based on two traces collected from an university campus. Our analysis shows that the proposed reordering approach for related list would lead to a 2 to 5 times increase in cache hit rate compared to an approach without reordering the related list. The increase in hit rate would lead to a 5.12% to 18.19% reduction in server load or back-end bandwidth usage. This increase in hit rate and reduction in back-end bandwidth reduces the latency in streaming the video requested by the viewer and has the potential to improve the overall performance of YouTube's content distribution system. An analysis of YouTube's recommendation system reveals that related lists are created from a small pool of videos, which increases the potential for caching content from related lists and reordering based on the content in the cache.

Efficient Implementation and Processing of a Real-Time Panorama Video Pipeline

Abstract: High resolution, wide field of view video generated from multiple camera feeds has many use cases. However, processing the different steps of a panorama video pipeline in real-time is challenging due to the high data rates and the stringent requirements of timeliness. We use panorama video in a sport analysis system where video events must be generated in real-time. In this respect, we present a system for real-time panorama video generation from an array of low-cost CCD HD video cameras. We describe how we have implemented different components and evaluated alternatives. We also present performance results with and without co-processors like graphics processing units (GPUs), and we evaluate each individual component and show how the entire pipeline is able to run in real-time on commodity hardware.

The Nornir run-time system for parallel programs using Kahn process networks on multi-core machines—a flexible alternative to MapReduce

Abstract: Even though shared-memory concurrency is a paradigm frequently used for developing parallel applications on small- and middle-sized machines, experience has shown that it is hard to use. This is largely caused by synchronization primitives which are low-level, inherently non-deterministic, and, consequently, non-intuitive to use. In this paper, we present the Nornir run-time system. Nornir is comparable to well-known frameworks such as MapReduce and Dryad that are recognized for their efficiency and simplicity. Unlike these frameworks, Nornir also supports process structures containing branches and cycles. Nornir is based on the formalism of Kahn process networks, which is a shared-nothing, message-passing model of concurrency. We deem this model a simple and deterministic alternative to shared-memory concurrency. Experiments with real and synthetic benchmarks on up to 8 CPUs show that performance in most cases scales almost linearly with the number of CPUs, when not limited by data dependencies. We also show that the modeling flexibility allows Nornir to outperform its MapReduce counterparts using well-known benchmarks.

Proceedings of the 4th ACM Multimedia Systems Conference

Abstract: We welcome you to the fourth ACM Multimedia Systems conference (MMSys). This conference focuses on multimedia computing and networking, lying at the intersection of digital media and the systems (applications, architectures, networking, operating systems) that handle them. The goal is to cut across a diverse set of computer science subfields in order to achieve a longitudinal perspective on the challenges that multimedia data types present. This provides a unique opportunity to view the intersections and interplay of the various approaches and solutions developed across these domains to deal with multimedia data types. Furthermore, MMSys provides an avenue for communicating research that addresses multimedia systems holistically.

Analysis of a real-world HTTP segment streaming case

Abstract: Today, adaptive HTTP segment streaming is a popular way to deliver video content to users. The benefits of HTTP segment streaming include its scalability, high performance and easy deployment, especially the possibility to reuse the already deployed HTTP infrastructure. However, current research focuses merely on client side statistics like for example achieved video qualities and adaption algorithms. To quantify the properties of such streaming systems from a service provider point of view, we have analyzed both sender and receiver side logging data provided by a popular Norwegian streaming provider. For example, we observe that more than 90% of live streaming clients send their requests for the same video segment with an inter-arrival time of only 10 seconds. Moreover, the logs indicate that the server sends substantially less data than is actually reported to be received by the clients, and the origin server streams data to less clients than there really are. Based on these facts, we conclude that HTTP segment streaming really makes use of the HTTP cache infrastructure without the need to change anything in the parts of Internet that are not controlled by the streaming provider.

Lecture de DCP pour le cinéma numérique avec le lecteur multimédia VLC et libav/ffmpeg

Abstract: Les salles de cinema sont passees a l’ere numerique. Pour la distribution des copies numeriques de films, la Digital Cinema Initiative (DCI) a choisi d’encoder les films au format Digital Cinema Package (DCP). Notre activite est menee en collaboration avec les cinemas Utopia, qui forment un reseau de salles independantes. La principale exigence d’Utopia est de fournir des logiciels libres et gratuits pour le cinema numerique. Le format DCP utilise la compression JPEG2000 pour la video, en raison de son taux de compression eleve pour les grandes images. Realiser une implementation efficace de codage et de decodage de ce format est complexe. Neanmoins, nous proposons une implementation amelioree du decodage pour la projection dans les salles obscures. Les equipements actuellement deployes dans les salles sont chers a l’achat et ont un cout de maintenance eleve, ce qui empeche les petites salles independantes de s’equiper. Notre but est de proposer une solution logicielle adaptee a la projection des DCP. Cet article presente une solution qui realise le decodage en temps reel et la projection, en s’appuyant sur des outils multimedia libres et standards comme le lecteur VLC et les bibliotheques libav/ffmpeg. Nous presentons les ameliorations implementees dans VLC pour supporter les DCP, qui incluent la lecture des fichiers et la synchronisation entre audio et video. Nous detaillons ensuite la realisation du decodeur JPEG2000 dans libav/ffmpeg. Pour finir, nous evaluons les performances de lecture atteintes.