Today, distributed applications run at data centre and Internet scales, from intensive data analysis, such as MapReduce; to the dynamic demands of a worldwide audience, such as YouTube. The network is essential to these applications at both scales. To provide adequate support, we must understand the full requirements of the applications, which are revealed by the workloads. In this thesis, we study distributed system applications at different scales to enrich this understanding. Large-scale Internet applications have been studied for years, such as social networking service (SNS), video on demand (VoD), and content delivery networks (CDN). An emerging type of video broadcasting on the Internet featuring crowdsourced live video streaming has garnered attention allowing platforms such as Twitch to attract over 1 million concurrent users globally. To better understand Twitch, we collected real-time popularity data combined with metadata about the contents and found the broadcasters rather than the content drives its popularity. Unlike YouTube and Netflix where content can be cached, video streaming on Twitch is generated instantly and needs to be delivered to users immediately to enable real-time interaction. Thus, we performed a large-scale measurement of Twitchs content location revealing the global footprint of its infrastructure as well as discovering the dynamic stream hosting and client redirection strategies that helped Twitch serve millions of users at scale. We next consider applications that run inside the data centre. Distributed computing applications heavily rely on the network due to data transmission needs and the scheduling of resources and tasks. One successful application, called Hadoop, has been widely deployed for Big Data processing. However, little work has been devoted to understanding its network. We found the Hadoop behaviour is limited by hardware resources and processing jobs presented. Thus, after characterising the Hadoop traffic on our testbed with a set of benchmark jobs, we built a simulator to reproduce Hadoops job traffic With the simulator, users can investigate the connections between Hadoop traffic and network performance without additional hardware cost. Different network components can be added to investigate the performance, such as network topologies, queue policies, and transport layer protocols. In this thesis, we extended the knowledge of networking by investigated two widelyused applications in the data centre and at Internet scale. We (i) studied the most popular live video streaming platform Twitch as a new type of Internet-scale distributed application revealing that broadcaster factors drive the popularity of such platform, and we (ii) discovered the footprint of Twitch streaming infrastructure and the dynamic stream hosting and client redirection strategies to provide an in-depth example of video streaming delivery occurring at the Internet scale, also we (iii) investigated the traffic generated by a distributed application by characterising the traffic of Hadoop under various parameters, (iv) with such knowledge, we built a simulation tool so users can efficiently investigate the performance of different network components under distributed application.

Over the last decade, live video streaming applications have gained great popularity among users but put great pressure on video servers and the Internet. In order to satisfy the growing demands for live video streaming, Peer-to-Peer(P2P) has been developed to relieve the video servers of bandwidth bottlenecks and computational load. Furthermore, Network Coding (NC) has been proposed and proved as a significant breakthrough in information theory and coding theory. According to previous research, NC not only brings substantial improvements regarding throughput and delay in data transmission, but also provides innovative solutions for multiple issues related to resource allocation, such as the coupon-collection problem, allocation and scheduling procedure. However, the complex NC-driven P2P streaming network poses substantial challenges to the packet scheduling algorithm. This thesis focuses on the packet scheduling algorithm for video multicast in NC-driven P2P streaming network. It determines how upload bandwidth resources of peer nodes are allocated in different transmission scenarios to achieve a better Quality of Service(QoS). First, an optimized rate allocation algorithm is proposed for scalable video transmission (SVT) in the NC-based lossy streaming network. This algorithm is developed to achieve the tradeoffs between average video distortion and average bandwidth redundancy in each generation. It determines how senders allocate their upload bandwidth to different classes in scalable data so that the sum of the distortion and the weighted redundancy ratio can be minimized. Second, in the NC-based non-scalable video transmission system, the bandwidth ineffi- ciency which is caused by the asynchronization communication among peers is reduced. First, a scalable compensation model and an adaptive push algorithm are proposed to reduce the unrecoverable transmission caused by network loss and insufficient bandwidth resources. Then a centralized packet scheduling algorithm is proposed to reduce the unin- formative transmission caused by the asynchronized communication among sender nodes. Subsequently, we further propose a distributed packet scheduling algorithm, which adds a critical scalability property to the packet scheduling model. Third, the bandwidth resource scheduling for SVT is further studied. A novel multiple- generation scheduling algorithm is proposed to determine the quality classes that the receiver node can subscribe to so that the overall perceived video quality can be maxi- mized. A single generation scheduling algorithm for SVT is also proposed to provide a faster and easier solution to the video quality maximization function. Thorough theoretical analysis is conducted in the development of all proposed algorithms, and their performance is evaluated via comprehensive simulations. We have demon- strated, by adjusting the conventional transmission model and involving new packet scheduling models, the overall QoS and bandwidth efficiency are dramatically improved. In non-scalable video streaming system, the maximum video quality gain can be around 5dB compared with the random push method, and the overall uninformative transmiss- sion ratio are reduced to 1% - 2%. In scalable video streaming system, the maximum video quality gain can be around 7dB, and the overall uninformative transmission ratio are reduced to 2% - 3%.

Submitted by Filipe dos Santos (fsantos@pucsp.br) on 2018-08-01T11:32:52Z No. of bitstreams: 1 Guilherme Henrique de Oliveira Cestari.pdf: 8168483 bytes, checksum: c1b8c3a5d071895d0a24c305e47a9b74 (MD5) / Made available in DSpace on 2018-08-01T11:32:52Z (GMT). No. of bitstreams: 1 Guilherme Henrique de Oliveira Cestari.pdf: 8168483 bytes, checksum: c1b8c3a5d071895d0a24c305e47a9b74 (MD5) Previous issue date: 2018-06-11 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / “Ambientation” is the continuous, purposive, and effective mediation of the multimodal qualities creating a space of signs. Together with djs, musicians, set and light designers as well as the participating public, video jockeys (vjs) compose dance floor ambientations, while manipulating and projecting audiovisual content in real time. This study adopts a semiotic approach in the inquiry into the strategies through which vjs create ambientations in their performances by means of live visuals and soundscapes. Nine parties, shows, and festivals from 2011 to 2016 are the object of detailed semiotic and multimedia analyses. Part 1 describes the data, exemplifies the variety of vjing scenarios, and identifies live performance, synaesthetic effects, and collage as the three principal strategies of creative practice. Part 2 introduces H. Gumbrecht’s theory of the production of presence in light of C. S. Peirce’s semiotics. It distinguishes six dimensions of presence that transform a dance floor into a creative semiotic space: (1) Stimmung (free possibilities of attunement), (2) mood (materialized quality), (3) atmosphere (or mood chaining), (4) the environment (movement in the atmospheres), (5) “scapes” (limitations and perceptual restrictions in the environment), and (6) ambience (patterns regulating atmospheric transitions). During a live video fruition, ambientation creates tensions oscillating between effects of presence (contact and rhythm) and effects of sense (intellectual depths and meanings). On the dance floor, vjs mobilize sensory qualities and combine them with actions and habits. They build a universe of Stimmungen from which all participants in the performance, immersed and surrounded in ambientation, are free to choose and create their own semiosic helix. The general effects of the event are commotion, emotive condition, and energetic reaction, together with effects of meaning / Ambientação é a mediação contínua, propositada e efetiva de qualidades multimodais que criam um espaço de signos. Juntamente com djs, músicos, cenógrafos, iluminadores e com o público participante, vídeo jóqueis (vjs) compõem ambientações nas pistas de dança ao manipular e projetar conteúdo audiovisual em tempo presente. Este estudo adota uma abordagem semiótica na investigação de estratégias usadas por vjs, que criam ambientações em suas performances por meio de visuais ao vivo e paisagens sonoras. Nove festas, shows e festivais ocorridos entre 2011 e 2016 são o objeto de análise semiótica detalhada. A seção I descreve os dados, exemplifica a variedade de cenários de vjing e identifica performance ao vivo, efeitos sinestésicos e colagem como três estratégias principais de prática criativa. A seção II apresenta a teoria sobre produção de presença de H. Gumbrecht à luz da semiótica de C. S. Peirce. Também distingue seis dimensões de presença que transformam a pista de dança em um espaço criativo de signos: (1) Stimmung (livres possibilidades de afinação), (2) humor (qualidade materializada), (3) atmosfera (ou encadeamento entre humores), (4) ambiente (movimento entre atmosferas), (5) paisagem (limitações e restrições perceptivas do ambiente) e (6) ambiência (padrões que regulam transições atmosféricas). Na pista de dança, vjs mobilizam qualidades sensórias, combinandoas com ações e hábitos. Eles constroem um universo de Stimmungen a partir dos quais todos os participantes na performance, imersos e rodeados por ambientação, se veem livres para escolher e criar sua própria hélice semiótica. Os efeitos gerais do evento são comoção, emotividade e reações energéticas, juntamente com efeitos de sentido

There are more and more people watching video streams over the Internet, and this has led to an increase in companies that compete for viewers. To improve the users experience, these companies can measure how their services are performing. The aim of this thesis was to recommend a way to measure the quality of service for a real time video streaming service. Three methods were presented; to buy the information from a content delivery network, extend existing analytics software or build a custom solution using packet sniffing. It was decided to extend existing analytics software. An evaluation was made on which software to extend. Four solutions were compared: Google Analytics, Mixpanel, Ooyala IQ and Piwik. The comparison was made using the analytic hierarchy process, comparing each alternative in their performance in criteria such as API maturity, flexibility, visualization and support. The recommended software to extend when building a real time video streaming service is Ooyala IQ which excel at flexibility and is easy to implement into existing solutions. It also had great capacity, offering no limit on how many events it can track per month, and finally it offers great dedicated support via telephone or email. / Det finns fler och fler personer som tittar på video strömmar på Internet, detta har lett till att nya företag har startats som konkurerar om tittare. För att förbättra kundupplevelsen kan man mäta hur tjänsten presterar. Målet med examensarbetet var att rekommendera hur man kan mäta tjänstekvalite för en realtidsvideoströmningstjänst. Tre olika lösningsförslag presenterades; att köpa informationen från en content delivery network, att bygga vidare på tillgängliga analytisk mjukvara eller att bygga ett eget paketsniffarprogram. Det bestämdes att bygga vidare på tillgänglig analytisk mjukvara. Fyra olika mjukvara jämfördes: Google Analytics, Mixpanel, Ooyala IQ och Piwik. Jämförelsen gjordes med hjälp av analytical hierarchy process, de olika alternativen jämfördes med avseende på: hur moget API:et var, flexibilitet, visualiseringen av data och support. Rekommendationen är att använda sig av Ooyala IQ som utmärker sig med avseende på flexibilitet, det var enkelt att använda deras API i sin egen lösning, det fanns ingen gräns på hur många händelser man kunde lagra per månad, och slutligen så fanns det dedikerad supportpersonal att nå via telefon eller email.

Šiame darbe išnagrinėtos virtualios mikrokontrolerio programavimo laboratorijos naudojamos kitose mokymo įstaigose, jų realizavimo būdai ir naudojamos technologijos. Palygintos technologijos ir jų galimybės. Išanalizuotos kontrolerio nuotolinio valdymo per internetinės svetainės sąsają, tiesioginio vaizdo perdavimo, vartotojų autentifikavimo, resursų planavimo, duomenų saugojimo realizacijos galimybės. Atsižvelgiant į tyrimų rezultatus, sukurta VL, ne brangesnė už šiuo metu naudojamą laboratorijos priemonių komplektą. Virtuali laboratorija leidžia pilnai realizuoti VDU dėstomų dalykų laboratorinius darbus, pasiekiama nepriklausomai nuo vietos ar laiko. Išspręsta resursų planavimo problema – panaudojant VL vartotojų registraciją bei laiko rezervacijas. Realizuota tiesioginė vaizdo transliacija, kuri palaikoma beveik visų interneto naršyklių. Virtualioje laboratorijoje yra realizuota dviejų analoginių įėjimų simuliacija ir indikacija, 4-ių skaitmeninių įėjimų/išėjimų indikacija, bei 3-jų skaitmeninių įėjimų/išėjimų simuliacija. Turi prijungtą LCD displėjų. VL talpinimui, panaudojus mini kompiuterį Raspberry Pi, ji vartoja tik 5 W energijos. Dėl pasirinktos standartinės kameros bei naudojamos PĮ, TĮ komponentai gali būti pakeisti, nesukeliant didesnių problemų. Raspberry Pi duomenys saugomi SD kortelėje, todėl pasidarius veikiančios sistemos kopiją, galima lengvai ir greitai sukurti kelias laboratorijas. / This paper deals with the investigation of the virtual microcontroller programming laboratories used in other educational institutions, as well as their realization techniques, and technologies used. Technologies and their offered potentials are compared. The possibilities of microcontroller remote control via website interface, live video streaming, user authentication, resource planning, and data storage are examined. Taking into account investigation results, the virtual laboratory has been developed, no more expensive than those currently used in laboratory work equipment kit per student. Virtual Lab allows to fully realize laboratory works of subjects taught at VDU, as it is fully available irrespective of space or time constraints. The issue of resource planning problem was solved using the VL user registration and time booking. Live video broadcasting has been implemented, which is supported by nearly all Internet browsers. Virtual laboratory has 4 digital PIN indicators and 2 analog inputs indicators. VL can simulate 2 analog inputs and 3 digital I/O. It has an LCD connected. VL is hosted using Raspberry Pi mini-computer – it uses only 5 watts of energy. The chosen standard camera and software used enables worn-out parts to be easily replaced while evading any problems. Raspberry Pi data stored on the SD card, thus by making a copy of the working VL, one can create multiple laboratories.

De plus en plus de metteurs en scène ont aujourd’hui recours à la vidéo en direct dans leurs spectacles. De manière générale, le recours à ce type de dispositifs pendant la représentation permet d’offrir au spectateur des possibilités de perception qu’il n’a habituellement pas au théâtre : voir le plus infime détail d’un visage, ou une action au plus près, accéder visuellement à tout espace extérieur à la scène lorsque l’action scénique s’y déroule, percevoir sur le visage d’un acteur sa réaction devant l’action qui se tient sous ses yeux ou encore avoir une multiplicité de points de vue simultanés sur un même objet. Pour autant, peut-on dire que la fonction des images en direct au théâtre se réduise à ces quelques principes utilitaires dans la représentation ? En donnant aux spectateurs la possibilité de voir simultanément les acteurs et leur image filmée et projetée et/ou diffusée en direct sur le plateau, les artistes jouent de plus en plus du potentiel performatif du dispositif vidéo dans la représentation théâtrale. Ils ne s’arrêtent pas à redoubler la scène par l’image mais mettent en avant la déconstruction de l’image et la construction du réel ou, du moins, d’une réalité dans et par l’image. C’est pourquoi le recours aux images en direct au théâtre appelle inévitablement à en interroger les enjeux à la fois esthétique, narratif, dramaturgique et proxémique dans la représentation théâtrale. Sans prétendre à une quelconque exhaustivité, cette thèse vise à proposer, à partir de certains spectacles d’Ivo van Hove et de Guy Cassiers, quelques pistes permettant de mesurer l’étendue possible du potentiel performatif des images en direct dans la représentation. / Today, more and more directors include live video in their performances. The use of such devices offers new possibilities of perception to the spectators, which usually are not available in a theatrical representation such as to look at the smallest details of an action or on the performer’s face; to access visually to peripheral areas of the stage when a dramatic action takes place there; to notice the reaction of an actor who listens to the main protagonist, as well as to have different simultaneous viewpoints on the same object. Nevertheless, can we reduce the function of live images on stage to these few utilitarian principles in the performance? Allowing spectators to see, in the same time, the actors and the images of themselves, many artists exploit more than ever the performative potential of video in theatrical representations. Live video does not only reproduce on the screen what happens on the scene, but it also highlights on one hand the deconstruction of the image and, on the other, the construction of a reality in and by images. That is why the uses of live video on stage unavoidably lead to question issues related to the aesthetic, narrative, dramaturgical and proxemic aspects of theatrical performances. Based on the analysis of different performances staged by Ivo van Hove and Guy Cassiers, this thesis aims to measure the extent of the performative potential of live video on stage.

The phenomenon of Live Video Shopping (LVS) has gained increased attention in recent years. Sinceapproximately 2017, the Chinese market has brought LVSto the attention of the public. In terms of overall market share, LVS was projected to account for roughly 20% of the overalle-commerce volume in Chinauntil 2021. Despite that, within Europe LVS has not seen such rapid growth and seems to be still a neglected domain,both in practice and in academia.Most academic research on LVS has originated in China and took place in an Asian setting. From a European standpoint, these findings can onlyserve as a starting point for further researchdue to cultural differences. Sincemost extant research on the topic of LVS adopted a quantitative research design, we could clearly identify a lack of qualitative studies in a European context. Therefore, our study aims to fill this research gap by investigating European consumersperceived value of LVS and their motivations to engage in it. A qualitative research design is particularly suitable for the exploratory nature of ourresearch. For the collection of data, we conducted three synchronous online focus groups through purposive sampling of European participants. In addition to that, and as a triangulation for our data sources, Pål Burman, the CEO of the LVS provider Zellma, was interviewed. With this we were able to includedifferent perspectives on the topic. It also helped us to increase thevalidityofthemanagerial implications.Usingthematic analysis, we could identify themes that theparticipants associated with the concept of LVS. The findings suggest that the perceived value of LVS for European consumers is not indisputable. Based on the theoretical concept of perceived value,LVS has to solve the trade-off between give and get components. This means that the committed time for attending an LVS stream has to be compensated by certainbenefits, such asdiscounts, enhanced product information, exclusivity of content,or inspiration. By using the theoretical framework of Uses and Gratifications Theory (UGT), we were able to divide the motivations of European consumers to engage in LVS in three types of gratifications: hedonic, utilitarian, and social. As opposed to prior Asian research on the topic, social gratifications only played a minor role forEuropean consumers. Reasons for an interaction between the viewer and the broadcaster were primarily product-related and utilitarian by nature. Engagement for the sake of social motives could not be confirmed. Utilitarian gratifications desired byEuropean consumers were mainly connected to theobtainment of enhanced production information, leading to more informed purchase decisions. Hedonic motivations were for exampleentertainmentand inspiration. When connecting UGT to the TechnologyAcceptance Model (TAM), we have also been able to identify consumers' approachesto LVS as a new technology.Crucial here weretheperceived ease of use and perceived usefulnessof LVS.It was shown that consumers require LVS to be as easy or easier than normal e-commerce shopping is already.LVS would further be perceived as useful if it could provide the viewers with something uniquethat old technologies are not able to provide.Asa new digital phenomenon, itis expected to serve as a complement to physical and online stores. In fact, LVS might not only serve as a platform of sales, but also tocreate long-lasting relationships between a brand and its consumers with consumer engagementas a main desired outcome.

Virtualization on top of high-performance servers has enabled the virtualization of network functions like caching, deep packet inspection, etc. Such Network Function Virtualization (NFV) is used to dynamically adapt to changes in network traffic and application popularity. We demonstrate how the combination of Software Defined Networking (SDN) and NFV can support the parallel operation of different Internet architectures on top of the same physical hardware. We introduce our architecture for this approach in an actual test setup, using CloudLab resources. We start of our evaluation in a small setup where we evaluate the feasibility of the SDN and NFV architecture and incrementally increase the complexity of the setup to run a live video streaming application. We use two vastly different protocol stacks, namely TCP/IP and NDN to demonstrate the capability of our approach. The evaluation of our approach shows that it introduces a new level of flexibility when it comes to operation of different Internet architectures on top of the same physical network and with this flexibility provides the ability to switch between the two protocol stacks depending on the application.

The live video streaming market is, as some research suggests, a difficult one to develop.Comparing it to other media it has the disadvantage of being available during a limitedperiod of time. This thesis explores how mobile devices can be used as an exploratorytool through live video streaming. The opportunity of control in the moment is studiedand examined in this report. This research will suggest a mobile device platform in whichpeople can assist other people exploring new environments around the globe. It also triesto evaluate whether this exploring interaction can occur between unacquainted people andat the same time not be hindered by any language barriers. The motivation behind thissubject is the human need and curiosity of exploring new environments and places.The foundation of the research approach was based on interaction design principles andmethods. This created a chain of methods used with the intent of better understandingwhat is to be considered when designing for live video experiences focusing onenvironments. The process lead to important understandings and key aspects to beconsidered when designing live experiences focusing on environments. The results thatemerged, showed that the most important encouragement is; to be presented with anenvironment that includes novel elements. Furthermore, the results could also concludethat by giving access to directional controls, and some means of expressing appreciation,can help to influence and shape exploration with live video.

Video streaming is widely recognized as the next Internet killer application. It was not one of the Internet's original target applications and its protocols (TCP in particular) were tuned mainly for e efficient bulk file transfer. As a result, a significant effort has focused on the development of UDP-based special protocols for streaming multimedia on the Internet. Recently, there has been a shift in video streaming from UDP to TCP, and specifically to HTTP. HTTP streaming provides a very attractive platform for video distribution on the Internet mainly because it can utilize all the current Internet infrastructure. In this thesis we make the argument that the marriage between HTTP streaming and the current Internet infrastructure can create many problems and inefficiencies. In order to solve these issues, we provide a set of techniques and protocols that can help both the network and end-hosts to make better decisions to improve video streaming quality. The thesis makes the following contributions: - We conduct a characterization study of popular commercial streaming services on mobile platforms. Our study shows that streaming services make different design decisions when implementing video players on different mobile platforms. We show that this can lead to several inefficiencies and undesirable behaviors specially when several clients compete for bandwidth in a shared bottleneck link. - Fairness between traffic flows has been preserved on the Internet through the use of TCP. However, due to the dynamics of adaptive video players and the lack of standard client adaptation techniques, fairness between multiple competing video flows is still an open issue of research. Our work extends the definition of standard bitrate fairness to utility fairness where utility is the Quality of Experience (QoE) of a video stream. We define QoE max-min fairness for a set of adaptive video flows competing for bandwidth in a network and we develop an algorithm that computes the set of bitrates that should be assigned to each stream to achieve fairness. We design and implement a system that can apply QoE fairness in home networks and evaluate the system on a real home router. - A well known problem that has been associated with TCP traffic is the buffer bloat problem. We use an experimental setup to show that adaptive video flows can cause buffer bloat which can significantly harm time sensitive applications sharing the same bottleneck link with video traffic. In addition, we develop a technique that can be used by video players to mitigate this problem. We implement our technique in a real video player and evaluate it on our testbed. - With the increasing popularity of video streaming on the Internet, the amounts of traffic on the peering links between video streaming providers and Internet Service Providers (ISPs) have become the source of many disputes. Hybrid CDN/P2P streaming systems can be used to reduce the amounts of traffic on the peering links by leveraging users upload bandwidth to redistribute some of the load to other peers. We develop an analysis for hybrid CDN/P2P systems that broadcast live adaptive video streams. The analysis helps the CDN to make better decisions to optimize video quality for its users.