On the Feasibility of DASH Streaming in the Cloud

Size: px
Start display at page:

Download "On the Feasibility of DASH Streaming in the Cloud"

Transcription

1 On the Feasibility of DASH Streaming in the Cloud Cong Wang and Michael Zink Department of Electrical and Computer Engineering University of Massachusetts, Amherst, MA 13 {cwang, ABSTRACT As shown in recent studies, video streaming is by far the biggest category of backbone Internet traffic in the US. As a measure to reduce the cost of highly overprovisioned physical infrastructures while remaining the quality of video services, many streaming service providers started to use cloud services where physical resources can be dynamically allocated based on current demand. This paper characterizes the performance of Dynamic Adaptive Streaming over HTTP (DASH), a new MPEG standard on adaptive streaming, in the cloud. We seek to answer the following questions that are critical to content providers that are hosting video in clouds: Which data center is the best to host videos? Does geographical distance matter? What type of instance is best suitable depending on different needs? How to efficiently solve the trade-off between performance and cost? The measurement methods and results presented in this paper can be easily expanded into other VoD services, and they allow us to i) characterize DASH behavior when streaming from the cloud; ii) identify the key factors that influence the DASH performance; and iii) suggest improvements for related services. Categories and Subject Descriptors C.4 [Performance of Systems]: Reliability, availability, and serviceability General Terms Performance, measurement Keywords Cloud computing, HTTP adaptive streaming, video-ondemand, quality of experience 1. INTRODUCTION Video Streaming in the Internet has recently been gaining tremendous popularity. In 211, YouTube had more than Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from Permissions@acm.org. NOSSDAV 14, March 19-21, 214, Singapore, Singapore Copyright 214 ACM /14/3...$ trillion views or around 14 views for every person on Earth [8]. According to [4], Netflix video sessions count for over 3% of Internet downstream traffic. The same report also shows that Netflix, YouTube, and Hulu together contribute for more than 5% of the Internet downstream traffic in North America. In addition, many traditional broadcast channels (e.g., NBC, ABC, CBS, etc.) now provide part of their programmes online on the web. Following this trend, the network and content providers have to supply a large amount of physical resources in order to meet the high demand of video services. As with many other services, resources are provided such that peak requests can be handled without impact on the quality of the provided VoD service; on the contrary, analyses of VoD services have shown that the overall number of requests can vary significantly over time. For example, in [21] we have shown that, YouTube requests follow a diurnal pattern, and peak demand lasts only for a relatively short period during a 24-hour interval. This can lead to significant overprovisioning of resources. As an alternative solution, cloud services, such as Amazon Web Services (AWS), provide resources that can be quickly deployed and released according to customer needs. This approach has been widely used for web and VoD services, such as Netflix, GitHub and Dropbox, since the content providers do not have to maintain expensive physical infrastructures. Instead, they only need to pay for the usage of the allocated cloud resources. Another trend in video streaming with increasing attention is Dynamic Adaptive Streaming over HTTP (DASH) [19, 18]. In this case, the quality of a video, along with resulting streaming bandwidth, can be adapted based on bottleneck bandwidth, server capacity, and client resources. Unlike traditional progressive streaming, DASH divides the videos into segments with each segment representing a portion of the video. In the progressive streaming approach, the client sends video request to the server, then establishes a continuous TCP connection for transmitting the whole video. In contrast, the DASH client periodically probes the available bandwidth and client buffer condition, then decides the next requested segment quality level. DASH provides high resilience to network variations and smooth playback experience, however, it requires much more interactions between the client and server and may increase the overhead. Typically, for a 6-minute video, in the case of YouTube progressive streaming, the client only needs to send 1 HTTP GET request containing requested video information. However, the YouTube adaptive streaming

2 (bitrates), while each quality layer is chopped into segments. The DASH client downloads the MPD in order to retrieve a complete list of components, including video, audio and subtitles. The DASH client then determines the preferred set of segments based on its capability (e.g., resolution), user preference (e.g., language), and network conditions (e.g., bandwidth or buffer status). And finally, the DASH client downloads and plays out the desired content. DASH videos are usually encoded with different segment lengths. Commercial approaches use segment lengths ranging from 2 seconds per fragment (e.g., Microsoft Smooth Streaming [5]) to 1 seconds per fragment (e.g., Apple HLS [2]). Figure 1: DASH Flow Diagram client sends 141 HTTP GET request, while a VLC-DASH client with segment length of 2 sends over 16 HTTP GET requests, depending on the frequency of quality switchings. Due to the two trends above, and currently there is no public information available regarding the performance of DASH-based streaming in the cloud, our goal is to provide an extensive evaluation of the performance of DASH video streaming from the cloud. In this paper, we investigate commercial (Amazon and Rackspace) and research (ExoGENI) clouds and stream from various network environments; also we evaluate the scalability of cloud VM nodes by running large scale distributed streaming experiments over PlanetLab [6]. Our findings include: i) the geographical location of cloud data centers is not the dominant factor for VoD streaming performance. E.g., the EC2 west (California) data center shows much better performance than EC2 east Virginia when serving clients located on the east coast (Massachusetts); ii) the video quality decreases dramatically when the number of parallel streaming sessions exceeds a certain threshold, and we show the threshold for each type of cloud instance; iii) Amazon EC2 cloud instances provide up to 7% higher throughput than Rackspace, however, it usually supports less parallel client connections. The remainder of this paper is outlined as follows. Section 2 provides a brief introduction on the operations of DASH. In Section 3, we describe the measured cloud services and our measurement setup. Section 4 presents the results from these measurements. In Section 5, we give a brief overview of related work. Section 6 concludes the paper and points out future works. 2. DYNAMIC ADAPTIVE STREAMING OVER HTTP (DASH) In the section, we provide a brief overview of the functionality of DASH, the recently developed MPEG adaptive streaming standard. DASH consists of two components: A Media Presentation Description (MPD) manifest file which contains the description of the available content, their URL addresses, available bit rates or qualities, segment lengths, and other characteristics; and video segments which contain the actual multimedia bit streams in the form of single or multiple files based on the encoder [13]. Figure 1 depicts the streaming flow process in DASH. In this case, a video is encoded into different quality layers 3. METHODOLOGY In this section, we give a brief overview on the cloud services that are used in our experiments, then we describe the metrics for evaluating the performance of DASH-based streaming in the cloud. 3.1 Measured Cloud Services Our measurement study was performed on two commercial cloud services: the AWS Elastic Compute Cloud (EC2) [1], Rackspace Cloud [7], and a research cloud service: ExoGENI [3]. Both EC2 and Rackspace provide resizable cloud computing capacity to execute applications on demand. The cost of commercial cloud services depends on the type of resources used and the duration of the usage, as well as additional factors such as the amount of I/O performed and the amount of storage used. In addition to regular cloud server hosting, the commercial clouds also offers services including cloud storage, CDN, cloud-based website hosting, etc. ExoGENI cloud is an open-source cloud platform, which allows users to programmatically manage a controllable, shared substrate on a reservation basis. The ExoGENI cloud uses a slice-based architecture which gives experimenters more flexibility than commercial clouds, since it allows them to create customized network topology for a compute cluster. 3.2 Experiment Setup We evaluate the DASH performance using instrumentalized VLC player version 2.1. equipped with DASH plugin [17]. For each of the cloud VMs, an Apache HTTP server is installed along with the MPD files, as well as the DASH segments of a video. We used the Big Buck Bunny animated video (about 1 minutes in length), which is divided into 2-second segments and encoded with 17 layers of bitrates varying from 1 to 1, kbps. Since the VLC media player cannot currently support H264/MP4 videos with dynamic resolution, we use video encoded at a constant resolution of p and a 24 fps frame rate. Our choice of 2-second segments is intended to provide high flexibility for quick adaption to bandwidth fluctuations. Meanwhile, we use a maximum buffer size of 3 seconds, to help the client smooth over temporary disruptions of connectivity. The videos are requested from three different data center locations of each cloud, nine origins in total. The nodes are located in Amazon EC2 cloud data centers in CA, OR and VA, Rackspace cloud data centers in IL, TX and VA, and ExoGENI cloud racks in NC, TX and MA, respectively. The type of cloud servers we used are shown in Table 1. Our three sets of experiments are described in detail in Section 4.

3 Cloud vcpus Memory (GB) Price ($/hr) EC Rackspace E-GENI Table 1: Summary of cloud server setup. 3.3 Measurement Metrics To best characterize the behavior of DASH streaming in the cloud, besides the traditional throughput measurements, we utilize additional evaluation metrics. In this section, we briefly introduce these metrics Bitrate The bitrate expresses the objective video quality that is actually received by the client. If there are enough segments (more than 3% of total buffer length, or 9 seconds) in the buffer, the player will choose to request the highest rate of those offered in the MPD file that is smaller than the bandwidth measured in the previous segment downloading process. Otherwise, the client player will request the lowest possible bitrate from the MPD file. Usually a higher bitrate means that a higher quality segment has been received by the client Video presentation quality Another video quality metric we considered is the presentation quality of video playback. Here we employ the metric proposed by Zink et al. [2], which provides a simple yet efficient way to determine the video quality from the viewer s perspective. The authors believe that two factors reduce the user experience: the video quality degradation and the frequency of quality switches. As shown in a comparison study, the spectrum reflects perceived quality better than PSNR. The overall spectrum of quality adaptive video v can be represented as: s(v) = T t=1 z t(h t 1 T i=1 T ( z jh j)) 2 (1) z j=1 i where the two terms h t and z t are defined as: h t: the number of layers in time slot t, t = 1,..., T ; z t: indication of a step in time slot t, z t {, 1}, t = 1,..., T. Usually, the higher the spectrum is, the more fluctuations occur while playing the video, thus it reflects lower video playback quality. The benefit of this metric is its ability to reflect the subjective quality observed by viewers, while it can be calculated very quickly Video initiation time In order to provide users with a smooth playback experience, media players usually buffer the initial part of the video before they actually start displaying the content. Most players, including VLC, which we use for the experiments presented in this paper, set a certain buffer level threshold. As soon as the buffer accumulates more than the threshold, the video playback starts. This initial waiting time (i.e., until the buffer reaches a certain level) depends on several factors, such as server load and network bandwidth. We observe this metric aiming to analyze whether the initial waiting time increases under certain load conditions for the video hosting VMs in the cloud. 4. EXPERIMENTAL RESULTS In this section, we present our experiment results. We start by characterizing the influence of geographical location of the data centers on DASH streaming. Second, we examine the scalability of cloud instances by performing parallel streaming sessions from a single VM to globally distributed clients. Finally, we look into the performance of different types of cloud VM instances, and the number of concurrent sessions that can be supported by these instance types. 4.1 DASH Performance with Cloud Services In this section, we characterize the DASH cloud performance by running single server, single client video streaming experiments from each of the cloud data centers described in Section 3. We stream the same video from each of the nine cloud sources, plus a reference video server located inside the UMass Amherst campus network. This set of experiments is repeated once per hour for a total duration of 24 hours. To avoid the variation of a single access network impairing the results, we repeat the above measurements from four different access networks that are commonly used by endusers: the UMass campus WiFi network, Amherst home Comcast cable Internet, town of Amherst public wireless Internet, and UMass campus WiMAX network. Since WiMAX network performance is sensitive to signal strength variation, we set up the experiment at a fixed point with ideal signal strength as mentioned in [11]. Average Playback Bitrate (kbps) Amherst Cloud Region Campus Wifi Campus WiMAX Home Wifi Public Wifi Figure 2: DASH average playback bitrate Figure 2 shows the average bitrate (out of 24 measurements) for each location, the error bars represent the standard deviation. For each data center location, the bitrates vary based on the access network conditions. Yet the bitrates still show very similar trends among different data center locations. The campus server shows the best performance among all the locations, while Amazon EC2 Virginia has the lowest bitrate. This demonstrates that DASH streaming can be improved when streaming from a server really close by, e.g., within the same LAN or region of the client. Yet beyond that, the performance is not strongly correlated to the geographical distance. Our measurement shows EC2 California actually performs better than EC2

4 Spectrum Amherst Cloud Region Campus Wifi Campus WiMAX Home Wifi Public Wifi Figure 3: DASH spectrum comparison Virginia. The performance difference among data centers is consistent with the results shown in [1]. We conjecture that EC2 Virginia s lower performance in this case is because it is the most heavily used one among Amazon s data centers, handling more than 8% of the total traffic generated by EC2. Here the ExoGENI research cloud shows comparable performance to the commercial clouds. Especially, EG-MA provides one of the highest bitrate and lowest spectra among the cloud nodes. In addition to the bitrates shown in Figure 2, the spectrum expresses the perceived quality for the viewers. The results in Figure 3 show a similar trend between the spectrum and the bitrate. The two figures above indicate that low bitrates are often correlated with more quality changes (high spectrum values) while high bitrate cases result in a better perceived quality. Yet exceptions exist: in the campus WiFi case, the bitrate is almost identical for EC2 OR and Rackspace TX. However, EC2 OR s spectrum is 3 units larger than that of Rackspace TX. This might be caused by a higher number of quality changes in the case of EC2 OR. 4.2 Scalability of DASH Cloud Streaming In this section, we examine the scalability of DASH streaming by requesting concurrent streaming sessions (clients) from each of the cloud VM. We make use of 6 geographically distributed PlanetLab nodes as streaming clients. This is to make sure that the local access network on the client side will not become the bottleneck. Starting with 2 parallel sessions, we increase the number of sessions to 1,8 by increments of 5. Using this method, we aim to observe how a single cloud VM performs when large amounts of video requests are getting in at the same time, and to determine the maximum number of parallel requests that can be served. In the case that more than 6 sessions are needed, we start multiple downloading sessions on each of the PlanetLab nodes, resulting in a maximum of 3 parallel sessions being active on each PlanetLab node. Since the behavior of a DASH streaming session is very similar to regular HTTP downloading, for simplicity, we run HTTP downloading of the video segments as the contention traffic from PlanetLab nodes, while the measurement is performed by running actual VLC streaming from a node connected to our campus Wifi network. We show the bitrate variation observed at the VLC client with increasing number of concurrent sessions in Figure 4. At the beginning, when the amount of contention traffic goes up, the bitrate slightly decreases yet overall remains stable. However, after a certain point, the bitrate drops dramatically, then quickly reaches, which means the video does not start playing at all. The initial buffering time (as percentage of the total video length of 1 minutes) is shown in Figure 5. For all cloud regions, the time starts from below 5% (3 seconds) and remains constant. It quickly increases to over 25% (15 seconds) after certain number of parallel sessions. As the number of concurrent sessions increases, less cloud instances are able to start playing the video. Therefore, for parallel sessions higher than 1 some bars are not shown. Both Figure 4 and Figure 5 show the same trend for each of the data centers: there exists a threshold for concurrent DASH connections, exceeding the threshold will cause low video quality and long buffering time. Therefore, it is useful to consider migrating load to either more cloud nodes or larger VMs early before the load actually reaches the threshold. In our measurements, the EC2 instances deliver higher bitrate when the number of concurrent sessions is low. However, their thresholds are lower than Rackspace s when both are using small instances. (Further comparisons of different types of cloud instances are presented in Section 4.3.) The ExoGENI cloud is for research purpose at no cost, and provides better scalability performance than commercial cloud when using comparable instance size, therefore, it offers a good platform for initially testing new cloud based streaming mechanisms. However, ExoGENI only allows the reservation of resources for short time periods, thus it is not suitable for hosting commercial, long running or large scale video services. Overall, we show that hosting DASH streaming servers in clouds is feasible from the perspective of scalability. Since in this section, the instances we have been using are relatively small, this represents the lower bound of cloud server performance. If larger scale video hosting is needed, the content providers can easily deploy higher capacity servers or increase the number of servers. Average Playback Bitrate (kbps) 1e+7 9e+6 8e+6 7e+6 6e+6 5e+6 4e+6 3e+6 2e+6 1e Number of Sessions Figure 4: The bitrate variation with number of planetlab sessions

5 Initiation Percentage of Whole Video Bandwidth (Mbps) AWS-EC2 Rackspace micro small medium large xlarge 2xlarge Number of Sessions Instance Type Figure 5: DASH video playback initiation time 4.3 Cloud Instance Capabilities Both EC2 and Rackspace offer customers the flexibility to choose different cloud instances at different prices. In this section, we investigate the DASH streaming performance on different types of cloud instances. We choose the EC2 CA and Rackspace TX data centers, since they perform better than others in the former experiment (as shown in Section 4.1). A brief summary of capabilities and prices of different instances is shown in Table 2. Amazon EC2 offers 7 types of general purpose instances varying from t1.micro to m3.2xlarge. We excluded the m3.xlarge instance because it is similar to m1.xlarge in the sense of vcpus and memory. Similarly, Rackspace provides 7 standard instance types, and we selected 6 of them that are comparable to the EC2 instances. For the micro instance, both EC2 and Rackspace share the same hourly price; other than that, the price of Rackspace instances is about twice as high as that of EC2. In return, Rackspace usually provides more vcpus, which might potentially influence the performance of applications when serving a large number of clients in parallel. In this section, we only compare between Amazon and Rackspace, because ExoGENI does not provide instance sizes that are comparable to all of those in the commercial clouds. Using the same method as mentioned in Section 4.2, we start our analysis by running 1 PlanetLab node downloads concurrently, then gradually increase the number of parallel downloading sessions, until the video on the VLC client fails to start. From this measurement, we aim to determine the maximum number of requests that can be handled by a single cloud VM type. Cloud Size vcpus Memory Price ($/hr) (GB) EC2/RS micro 1/1.6/.5.2/.2 EC2/RS small 1/1 1.7/2.6/.12 EC2/RS medium 1/2 3.7/4.12/.24 EC2/RS large 2/4 7.5/8.24/.48 EC2/RS xlarge 4/6 15/15.48/.9 EC2/RS 2xlarge 4/8 3/3.9/1.2 Table 2: Summary of EC2 and Rackspace instances. Figure 6 shows the maximum bandwidth provided by each type of cloud instance. As can be observed, EC2 has up to 7% higher peak bandwidth than Rackspace. This might be the cause for EC2-CA showing a better performance than Figure 6: Maximum bandwidth of each type of cloud instance Number of Sessions AWS-EC2 Rackspace micro small medium large Instance Type xlarge 2xlarge Figure 7: Parallel sessions supported by each type of cloud instance Spectrum AWS-EC2 Rackspace micro small medium large Instance Type xlarge 2xlarge Figure 8: The spectrum condition when instance running near maximum number of parallel sessions Rackspace-TX in the experiment presented in Section 4.1. Interestingly, Figure 7 shows that Rackspace can support 3% more parallel sessions than EC2. Our conjecture is that Rackspace provides more vcpus for the same instance type, which can better handle large amounts of HTTP requests at the same time. This observation is consistent with the results presented in [12], which shows that most of the

6 Rackspace instances perform better than Amazon instances when processing parallel tasks. Figure 8 shows the spectrum for the VLC player session at the point when each type of instance is running close to the maximum number of parallel sessions. This figure demonstrates that, while larger instances can support more parallel sessions and a higher cumulative bandwidth, the quality of the individual streams reduces. Between an EC2 micro and 2xlarge instance the spectrum increases significantly from 3,2 to 3,9. Thus, in the case when server operates near the threshold, it might be more beneficial from a viewer s perspective to stream from a large set of small instances instead of a single large one. We propose to further investigate our claim in future experiments by comparing the results of streaming from multiple small instances in parallel. 5. RELATED WORK There has been extensive works on the performance evaluation of both Amazon and Rackspace cloud services. [14] provides a benchmark to compare cloud services in the sense of CPU, memory, network, etc; while [1] presents results from network measurements focusing on multiple cloud services provided by AWS, such as EC2, S3 and CloudFront. Works such as [15] present studies on cloud VM starting time and other metrics. On the DASH side, many works have been focusing on the DASH algorithm efficiency and encoding schemes, such as [13] and [9]; while others, such as [16], provides evaluation of DASH in different environment settings. However, these works are mostly done in a relatively small scale environment. To the best of our knowledge, our work is the first one that systematically investigates the effectiveness of DASH adaptive streaming in the cloud based on real distributed system measurements. 6. CONCLUSIONS In this paper, we evaluate the effectiveness of hosting adaptive streaming video servers in the cloud. Through a measurement study, we demonstrate the cloud resource and network performance of DASH streaming from the cloud. We characterize the DASH performance in nine data centers of three cloud service providers. In particular, we study the impact of geographical distance on DASH streaming, as well as the threshold for the number of concurrent sessions that can be hosted by a single cloud VM node; also we look into the methods of choosing best suitable cloud resources based on these thresholds. In future work, we intend to extend our work to the performance comparison among different DASH implementations, e.g., YouTube adaptive streaming and DASH-JS, the Javascript DASH library. Also we intend to further investigate the issues we identified in this paper, such as adaptive video delivery that is tolerant to network instability and the trade-off between single large cloud instance and multiple small ones. Acknowledgement This work is funded by and NSF grant CNS Any opinions, findings, conclusions or recommendations expressed in this material are the authors and do not necessarily reflect the views of the National Science Foundation. 7. REFERENCES [1] Amazon s Elastic Compute Cloud. [2] Apple HTTP Live Streaming. [3] ExoGENI. [4] Global internet phenomena report. sandvine.com/news/global_broadband_trends.asp. [5] Microsoft Smooth Streaming. [6] PlanetLab. [7] Rackspace Cloud. [8] YouTube Statistics. [9] S. Akhshabi, A. C. Begen, and C. Dovrolis. An experimental evaluation of rate-adaptation algorithms in adaptive streaming over HTTP. In MMSys, pages , 211. [1] I. Bermudez, S. Traverso, M. Mellia, and M. Munafo. Exploring the cloud from passive measurements: The Amazon AWS case. In INFOCOM, pages , 213. [11] F. Fund, C. Wang, T. Korakis, M. Zink, and S. Panwar. GENI WiMAX performance: Evaluation and comparison of two campus testbeds. In Proceedings of the GENI Research and Educational Experiment Workshop, 213. [12] S. Kalpit, S. S., and R. S. Cloud performance benchmark series. Cloud Commons Online, 211. [13] S. Lederer, C. Müller, and C. Timmerer. Dynamic Adaptive Streaming over HTTP Dataset. In MMSys, pages 89 94, 212. [14] A. Li, X. Yang, S. Kandula, and M. Zhang. CloudCmp: Comparing public cloud providers. In IMC, pages 1 14, 21. [15] M. Mao and M. Humphrey. A performance study on the VM startup time in the cloud. In Cloud Computing (CLOUD), pages , 212. [16] C. Mueller, S. Lederer, and C. Timmerer. An Evaluation of Dynamic Adaptive Streaming over HTTP in Vehicular Environments. In MoVid, February 212. [17] C. Müller and C. Timmerer. A VLC media player plugin enabling dynamic adaptive streaming over HTTP. In MMSys, pages , 211. [18] T. Stockhammer. Dynamic Adaptive Streaming over HTTP : Standards and Design Principles. In MMSys, pages , 211. [19] C. Timmerer and C. Müller. HTTP Streaming of MPEG Media. In Streaming Day, September 21. [2] M. Zink, J. Schmitt, and R. Steinmetz. Layer-encoded video in scalable adaptive streaming. IEEE Transactions on Multimedia, 7(1):75 84, 25. [21] M. Zink, K. Suh, Yu, and J. Kurose. Characteristics of YouTube Network Traffic at a Campus Network - Measurements, Models, and Implications. Elsevier Computer Networks, 29.

On the Feasibility of Prefetching and Caching for Online TV Services: A Measurement Study on

On the Feasibility of Prefetching and Caching for Online TV Services: A Measurement Study on See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/220850337 On the Feasibility of Prefetching and Caching for Online TV Services: A Measurement

More information

A QoE Friendly Rate Adaptation Method for DASH

A QoE Friendly Rate Adaptation Method for DASH A QoE Friendly Rate Adaptation Method for DASH Yuming Cao 1,3, Xiaoquan You 2,3, Jia Wang 1,3, Li Song 1,3 1 Institute of Image Communication and Network Engineering, Shanghai Jiao Tong University 2 Communication

More information

QoS Featured Wireless Virtualization based on Hardware

QoS Featured Wireless Virtualization based on Hardware QoS Featured Wireless Virtualization based on 802.11 Hardware Cong Wang and Michael Zink Department of Electrical and Computer Engineering University of Massachusetts, Amherst, MA 01003 {cwang, zink} @ecs.umass.edu

More information

IMPROVING LIVE PERFORMANCE IN HTTP ADAPTIVE STREAMING SYSTEMS

IMPROVING LIVE PERFORMANCE IN HTTP ADAPTIVE STREAMING SYSTEMS IMPROVING LIVE PERFORMANCE IN HTTP ADAPTIVE STREAMING SYSTEMS Kevin Streeter Adobe Systems, USA ABSTRACT While HTTP adaptive streaming (HAS) technology has been very successful, it also generally introduces

More information

A Tale of Three CDNs

A Tale of Three CDNs A Tale of Three CDNs An Active Measurement Study of Hulu and Its CDNs Vijay K Adhikari 1, Yang Guo 2, Fang Hao 2, Volker Hilt 2, and Zhi-Li Zhang 1 1 University of Minnesota - Twin Cities 2 Bell Labs,

More information

Internet Video Delivery. Professor Hui Zhang

Internet Video Delivery. Professor Hui Zhang 18-345 Internet Video Delivery Professor Hui Zhang 1 1990 2004: 1 st Generation Commercial PC/Packet Video Technologies Simple video playback, no support for rich app Not well integrated with Web browser

More information

An Experimental Evaluation of Rate Adaptation Algorithms in Adaptive Streaming over HTTP

An Experimental Evaluation of Rate Adaptation Algorithms in Adaptive Streaming over HTTP An Experimental Evaluation of Rate Adaptation Algorithms in Adaptive Streaming over HTTP Saamer Akhshabi, Constantine Dovrolis Georgia Institute of Technology Ali C. Begen Cisco Systems February 24, 2011

More information

Characterizing Netflix Bandwidth Consumption

Characterizing Netflix Bandwidth Consumption Characterizing Netflix Bandwidth Consumption Dr. Jim Martin Associate Professor School of Computing Clemson University jim.martin@cs.clemson.edu http://www.cs.clemson.edu/~jmarty Terry Shaw Director, Network

More information

MPEG's Dynamic Adaptive Streaming over HTTP - An Enabling Standard for Internet TV. Thomas Stockhammer Qualcomm Incorporated

MPEG's Dynamic Adaptive Streaming over HTTP - An Enabling Standard for Internet TV. Thomas Stockhammer Qualcomm Incorporated MPEG's Dynamic Adaptive Streaming over HTTP - An Enabling Standard for Internet TV Thomas Stockhammer Qualcomm Incorporated ABSTRACT Internet video is experiencing a dramatic growth in both fixed and mobile

More information

A Comparative Case Study of HTTP Adaptive Streaming Algorithms in Mobile Networks

A Comparative Case Study of HTTP Adaptive Streaming Algorithms in Mobile Networks Theodoros Karagkioules, Cyril Concolato, Dimitrios Tsilimantos and Stefan Valentin LTCI Telecom ParisTech Universite Paris-Saclay Mathematical and Algorithmic Sciences Lab France Research Center Huawei

More information

Future Buffer based Adaptation for VBR Video Streaming over HTTP

Future Buffer based Adaptation for VBR Video Streaming over HTTP Future Buffer based Adaptation for VBR Video Streaming over HTTP Tuan Vu 1, Hung T. Le 2, Duc V. Nguyen 2, Nam Pham Ngoc 1, Truong Cong Thang 2 1 Hanoi University of Science and Technology, Hanoi, Vietnam

More information

Page 1. Outline / Computer Networking : 1 st Generation Commercial PC/Packet Video Technologies

Page 1. Outline / Computer Networking : 1 st Generation Commercial PC/Packet Video Technologies Outline 15-441/15-641 Computer Networking Lecture 18 Internet Video Delivery Peter Steenkiste Slides by Professor Hui Zhang Background Technologies: - HTTP download - Real-time streaming - HTTP streaming

More information

What Happens When HTTP Adaptive Streaming Players Compete for Bandwidth?

What Happens When HTTP Adaptive Streaming Players Compete for Bandwidth? What Happens When HTTP Adaptive Streaming Players Compete for Bandwidth? Saamer Akhshabi, Ali C. Begen Lakshmi Anantakrishnan, Video and Content Platforms Research and Constantine Dovrolis, Advanced Development

More information

Characterizing Netflix Bandwidth Consumption

Characterizing Netflix Bandwidth Consumption Characterizing Netflix Bandwidth Consumption Jim Martin, Yunhui Fu, Nicholas Wourms School of Computing, Clemson University, Clemson, SC 93 {jmarty,yfu,nwourms}@clemson.edu Terry Shaw Cable Television

More information

QoE-Driven Video Streaming and Video Content Caching

QoE-Driven Video Streaming and Video Content Caching CommNet2 & IcoreJoint Workshop on Content Caching & Distributed Storage for Future Communication Networks QoE-Driven Video Streaming and Video Content Caching Xiaohong Peng Adaptive Communications Networks

More information

Connecting to the Cloud

Connecting to the Cloud Connecting to the Cloud Tomorrow is Now: What we are Connecting to the Cloud Robert Stevenson Chief Business Officer & SVP Strategy GAIKAI, a Sony Computer Entertainment Company DCIA Conference at CES

More information

Optimized Strategies for Real-Time Multimedia Communications from Mobile Devices

Optimized Strategies for Real-Time Multimedia Communications from Mobile Devices Optimized Strategies for Real-Time Multimedia Communications from Mobile Devices Enrico Masala Dept. of Control and Computer Engineering, Politecnico di Torino, Torino, Italy ( Part of this work has been

More information

Adaptive Video Acceleration. White Paper. 1 P a g e

Adaptive Video Acceleration. White Paper. 1 P a g e Adaptive Video Acceleration White Paper 1 P a g e Version 1.0 Veronique Phan Dir. Technical Sales July 16 th 2014 2 P a g e 1. Preface Giraffic is the enabler of Next Generation Internet TV broadcast technology

More information

Insights into PPLive: A Measurement Study of a Large-Scale P2P IPTV System

Insights into PPLive: A Measurement Study of a Large-Scale P2P IPTV System Insights into PPLive: A Measurement Study of a Large-Scale P2P IPTV System Xiaojun Hei, Chao Liang, Jian Liang, Yong Liu and Keith W. Ross Department of Computer and Information Science Department of Electrical

More information

SARA: Segment Aware Rate Adaptation for DASH Video Services

SARA: Segment Aware Rate Adaptation for DASH Video Services SARA: Segment Aware Rate Adaptation for DASH Video Services, Venkatesh Tamarapalli*, Deep Medhi University of Missouri Kansas City * Indian Institute of Technology-Guwahati, India Overview Introduction

More information

Datasets for AVC (H.264) and HEVC (H.265) for Evaluating Dynamic Adaptive Streaming over HTTP (DASH)

Datasets for AVC (H.264) and HEVC (H.265) for Evaluating Dynamic Adaptive Streaming over HTTP (DASH) Datasets for AVC (H.264) and HEVC (H.265) for Evaluating Dynamic Adaptive Streaming over HTTP (DASH) Jason J. Quinlan, Ahmed H. Zahran, Cormac J. Sreenan Dept. of Computer Science, University College Cork

More information

SamKnows test methodology

SamKnows test methodology SamKnows test methodology Download and Upload (TCP) Measures the download and upload speed of the broadband connection in bits per second. The transfer is conducted over one or more concurrent HTTP connections

More information

Multimedia Streaming. Mike Zink

Multimedia Streaming. Mike Zink Multimedia Streaming Mike Zink Technical Challenges Servers (and proxy caches) storage continuous media streams, e.g.: 4000 movies * 90 minutes * 10 Mbps (DVD) = 27.0 TB 15 Mbps = 40.5 TB 36 Mbps (BluRay)=

More information

Reduction of Periodic Broadcast Resource Requirements with Proxy Caching

Reduction of Periodic Broadcast Resource Requirements with Proxy Caching Reduction of Periodic Broadcast Resource Requirements with Proxy Caching Ewa Kusmierek and David H.C. Du Digital Technology Center and Department of Computer Science and Engineering University of Minnesota

More information

QoE-aware Traffic Shaping for HTTP Adaptive Streaming

QoE-aware Traffic Shaping for HTTP Adaptive Streaming , pp.33-44 http://dx.doi.org/10.14257/ijmue.2014.9.2.04 QoE-aware Traffic Shaping for HTTP Adaptive Streaming Xinying Liu 1 and Aidong Men 2 1,2 Beijing University of Posts and Telecommunications No.10

More information

Empirical Evaluation of Latency-Sensitive Application Performance in the Cloud

Empirical Evaluation of Latency-Sensitive Application Performance in the Cloud Empirical Evaluation of Latency-Sensitive Application Performance in the Cloud Sean Barker and Prashant Shenoy University of Massachusetts Amherst Department of Computer Science Cloud Computing! Cloud

More information

Dynamic Adaptive Streaming over HTTP (DASH) Application Protocol : Modeling and Analysis

Dynamic Adaptive Streaming over HTTP (DASH) Application Protocol : Modeling and Analysis Dynamic Adaptive Streaming over HTTP (DASH) Application Protocol : Modeling and Analysis Dr. Jim Martin Associate Professor School of Computing Clemson University jim.martin@cs.clemson.edu http://www.cs.clemson.edu/~jmarty

More information

AWS Elemental MediaPackage. User Guide

AWS Elemental MediaPackage. User Guide AWS Elemental MediaPackage User Guide AWS Elemental MediaPackage: User Guide Copyright 2018 Amazon Web Services, Inc. and/or its affiliates. All rights reserved. Amazon's trademarks and trade dress may

More information

Video Quality for Live Adaptive Bit-Rate Streaming: Achieving Consistency and Efficiency

Video Quality for Live Adaptive Bit-Rate Streaming: Achieving Consistency and Efficiency Video Quality for Live Adaptive Bit-Rate Streaming: Achieving Consistency and Efficiency Introduction The video industry is undergoing an unprecedented amount of change. More premium live video content

More information

Achieving Low-Latency Streaming At Scale

Achieving Low-Latency Streaming At Scale Achieving Low-Latency Streaming At Scale Founded in 2005, Wowza offers a complete portfolio to power today s video streaming ecosystem from encoding to delivery. Wowza provides both software and managed

More information

Quantifying Internet End-to-End Route Similarity

Quantifying Internet End-to-End Route Similarity Quantifying Internet End-to-End Route Similarity Ningning Hu and Peter Steenkiste Carnegie Mellon University Pittsburgh, PA 523, USA {hnn, prs}@cs.cmu.edu Abstract. Route similarity refers to the similarity

More information

irtc: Live Broadcasting

irtc: Live Broadcasting 1 irtc: Live Broadcasting Delivering ultra-low-latency media at massive scale with LiveSwitch and WebRTC Introduction In the early days of the Internet and personal computing, it wasn t uncommon to wait

More information

Automated Performance Evaluation of Adaptive HTML5 Player Deployments

Automated Performance Evaluation of Adaptive HTML5 Player Deployments Automated Performance Evaluation of Adaptive HTML5 Player Deployments Anatoliy Zabrovskiy, Evgeny Petrov, Evgeny Kuzmin Petrozavodsk State University Petrozavodsk, Russia {z anatoliy, johnp, kuzmin}@petrsu.ru

More information

ADVANCED TRANSPORT OPTIONS FOR THE DYNAMIC ADAPTIVE STREAMING OVER HTTP. Christian Timmerer, and Alan Bertoni

ADVANCED TRANSPORT OPTIONS FOR THE DYNAMIC ADAPTIVE STREAMING OVER HTTP. Christian Timmerer, and Alan Bertoni ADVANCED TRANSPORT OPTIONS FOR THE DYNAMIC ADAPTIVE STREAMING OVER HTTP Christian Timmerer, and Alan Bertoni Alpen-Adria-Universität Klagenfurt, Institute of Information Technology (ITEC), Austria {firsname.lastname}@itec.aau.at

More information

Low Latency MPEG-DASH System over HTTP 2.0 and WebSocket

Low Latency MPEG-DASH System over HTTP 2.0 and WebSocket Low Latency MPEG-DASH System over HTTP 2.0 and WebSocket Xiaona Wu 1,2, Cheng Zhao 1, Rong Xie 1,2, and Li Song 1,2 1 Institute of Image Communication and Network Engineering, Shanghai Jiao Tong University

More information

Towards Peer-Assisted Dynamic Adaptive Streaming over HTTP

Towards Peer-Assisted Dynamic Adaptive Streaming over HTTP Towards Peer-Assisted Dynamic Adaptive Streaming over HTTP Stefan Lederer, Christopher Müller, and Christian Timmerer Alpen-Adria-Universität Klagenfurt Universitätsstraße 65-67 9020 Klagenfurt am Wörthersee,

More information

Emulation of Dynamic Adaptive Streaming over HTTP with Mininet

Emulation of Dynamic Adaptive Streaming over HTTP with Mininet Emulation of Dynamic Adaptive Streaming over HTTP with Mininet Anatoliy Zabrovskiy Evgeny Kuzmin Petrozavodsk State University Video streaming Video streaming is becoming more and more popular technology

More information

Guaranteeing Video Quality

Guaranteeing Video Quality Guaranteeing Video Quality in IP Delivery Systems By Göran Appelquist, Ph.D., Chief Technology Officer, Edgeware AB This article explores some of the challenges and solutions for operators to guarantee

More information

Video Splicing Techniques for P2P Video Streaming

Video Splicing Techniques for P2P Video Streaming Video Splicing Techniques for P2P Video Streaming MD Iftakharul Islam and Javed I Khan Kent State University Email: mislam4@kent.edu, javed@cs.kent.edu Abstract In HTTP live streaming (HLS), a video is

More information

To Chunk or Not to Chunk: Implications for HTTP Streaming Video Server Performance

To Chunk or Not to Chunk: Implications for HTTP Streaming Video Server Performance To Chunk or Not to Chunk: Implications for HTTP Streaming Video Server Performance Jim Summers, Tim Brecht University of Waterloo jasummer, brecht@cs.uwaterloo.ca Derek Eager University of Saskatchewan

More information

ELECTRONIC COPY SAMKNOWS ANALYSIS OF ROGERS BROADBAND PERFORMANCE IN FEBRUARY 2015 ELECTRONIC COPY. Delivered by to: Shane Jansen.

ELECTRONIC COPY SAMKNOWS ANALYSIS OF ROGERS BROADBAND PERFORMANCE IN FEBRUARY 2015 ELECTRONIC COPY. Delivered by  to: Shane Jansen. ELECTRONIC COPY SAMKNOWS ANALYSIS OF ROGERS BROADBAND PERFORMANCE IN FEBRUARY 2015 Delivered by Email to: Shane Jansen Rogers Dated: February 25, 2015 ELECTRONIC COPY [THIS PAGE LEFT INTENTIONALLY BLANK]

More information

QoE-driven Dynamic Adaptive Video Streaming Strategy with Future Information

QoE-driven Dynamic Adaptive Video Streaming Strategy with Future Information 1 QoE-driven Dynamic Adaptive Video Streaming Strategy with Future Information Li YU, Student Member, IEEE, Tammam TILLO, Senior Member, IEEE, and Jimin XIAO, Member, IEEE Abstract Dynamic Adaptive Video

More information

TripleStream Product Description Version 4.6

TripleStream Product Description Version 4.6 TripleStream Product Description Version 4.6 Tripleplay Services Ltd. Rapier House 40-46 Lamb's Conduit Street London WC1N 3LJ www.tripleplay-services.com 2014 Tripleplay Services Ltd. All rights reserved.

More information

Abstract Video streaming has been rising for that past years and is responsible for the majority of all consumer Internet traffic.

Abstract Video streaming has been rising for that past years and is responsible for the majority of all consumer Internet traffic. Bachelor Informatica Informatica Universiteit van Amsterdam Performance Analysis of Virtualized Video Streaming Service Mark Croes Student number: 10609059 June 8, 2017 Supervisor(s): Dirk Griffioen (Unified

More information

Azure database performance Azure performance measurements February 2017

Azure database performance Azure performance measurements February 2017 dbwatch report 1-2017 Azure database performance Azure performance measurements February 2017 Marek Jablonski, CTO dbwatch AS Azure database performance Introduction The popular image of cloud services

More information

Anatomy of a DASH Client. Ali C. Begen, Ph.D.

Anatomy of a DASH Client. Ali C. Begen, Ph.D. Anatomy of a DASH Client Ali C. Begen, Ph.D. http://ali.begen.net Video Delivery over HTTP Enables playback while still downloading Server sends the file as fast as possible Pseudo Streaming Enables seeking

More information

A Seamless Web Integration of Adaptive HTTP Streaming

A Seamless Web Integration of Adaptive HTTP Streaming 20th European Signal Processing Conference (EUSIPCO 2012) Bucharest, Romania, August 27-31, 2012 A Seamless Web Integration of Adaptive HTTP Streaming Benjamin Rainer, Stefan Lederer, Christopher Müller,

More information

Hybrid Auto-scaling of Multi-tier Web Applications: A Case of Using Amazon Public Cloud

Hybrid Auto-scaling of Multi-tier Web Applications: A Case of Using Amazon Public Cloud Hybrid Auto-scaling of Multi-tier Web Applications: A Case of Using Amazon Public Cloud Abid Nisar, Waheed Iqbal, Fawaz S. Bokhari, and Faisal Bukhari Punjab University College of Information and Technology,Lahore

More information

Interactive Branched Video Streaming and Cloud Assisted Content Delivery

Interactive Branched Video Streaming and Cloud Assisted Content Delivery Interactive Branched Video Streaming and Cloud Assisted Content Delivery Niklas Carlsson Linköping University, Sweden @ Sigmetrics TPC workshop, Feb. 2016 The work here was in collaboration... Including

More information

Lecture 27 DASH (Dynamic Adaptive Streaming over HTTP)

Lecture 27 DASH (Dynamic Adaptive Streaming over HTTP) CS 414 Multimedia Systems Design Lecture 27 DASH (Dynamic Adaptive Streaming over HTTP) Klara Nahrstedt Spring 2012 Administrative MP2 posted MP2 Deadline April 7, Saturday, 5pm. APPLICATION Internet Multimedia

More information

Adaptation Algorithm for Adaptive Streaming over HTTP

Adaptation Algorithm for Adaptive Streaming over HTTP Adaptation Algorithm for Adaptive Streaming over HTTP Konstantin Miller, Emanuele Quacchio, Gianluca Gennari and Adam Wolisz Technische Universität Berlin, Germany Email: {konstantin.miller, adam.wolisz}@tu-berlin.de

More information

Performance Evaluation of Video Streaming in Multihop Wireless Mesh Networks

Performance Evaluation of Video Streaming in Multihop Wireless Mesh Networks Performance Evaluation of Video Streaming in Multihop Wireless Mesh Networks Xiaolin Cheng Prasant Mohapatra Sung-Ju Lee Sujata Banerjee Department of Computer Science, University of California at Davis,

More information

QoE Characterization for Video-On-Demand Services in 4G WiMAX Networks

QoE Characterization for Video-On-Demand Services in 4G WiMAX Networks QoE Characterization for Video-On-Demand Services in 4G WiMAX Networks Amitabha Ghosh IBM India Research Laboratory Department of Electrical Engineering University of Southern California, Los Angeles http://anrg.usc.edu/~amitabhg

More information

Watching User Generated Videos with Prefetching

Watching User Generated Videos with Prefetching Watching User Generated Videos with Prefetching Samamon Khemmarat, Renjie Zhou, Lixin Gao, Michael Zink Electrical and Computer Engineering Department UMass Amherst MMSys 2011, Feb. 24 th User Generated

More information

It s Not the Cost, It s the Quality! Ion Stoica Conviva Networks and UC Berkeley

It s Not the Cost, It s the Quality! Ion Stoica Conviva Networks and UC Berkeley It s Not the Cost, It s the Quality! Ion Stoica Conviva Networks and UC Berkeley 1 A Brief History! Fall, 2006: Started Conviva with Hui Zhang (CMU)! Initial goal: use p2p technologies to reduce distribution

More information

Complex Interactions in Content Distribution Ecosystem and QoE

Complex Interactions in Content Distribution Ecosystem and QoE Complex Interactions in Content Distribution Ecosystem and QoE Zhi-Li Zhang Qwest Chair Professor & Distinguished McKnight University Professor Dept. of Computer Science & Eng., University of Minnesota

More information

RealMedia Streaming Performance on an IEEE b Wireless LAN

RealMedia Streaming Performance on an IEEE b Wireless LAN RealMedia Streaming Performance on an IEEE 802.11b Wireless LAN T. Huang and C. Williamson Proceedings of IASTED Wireless and Optical Communications (WOC) Conference Banff, AB, Canada, July 2002 Presented

More information

Multimedia: video ... frame i+1

Multimedia: video ... frame i+1 Multimedia: video video: sequence of images displayed at constant rate e.g. 24 images/sec digital image: array of pixels each pixel represented by bits coding: use redundancy within and between images

More information

Active Adaptation in QoS Architecture Model

Active Adaptation in QoS Architecture Model Active Adaptation in QoS Architecture Model Drago agar and Snjeana Rimac -Drlje Faculty of Electrical Engineering University of Osijek Kneza Trpimira 2b, HR-31000 Osijek, CROATIA Abstract - A new complex

More information

Watch Global, Cache Local: YouTube Network Traffic at a Campus Network - Measurements and Implications

Watch Global, Cache Local: YouTube Network Traffic at a Campus Network - Measurements and Implications University of Massachusetts Amherst ScholarWorks@UMass Amherst Computer Science Department Faculty Publication Series Computer Science 2008 Watch Global, Cache Local: YouTube Network Traffic at a Campus

More information

Steady stream of bits 9/6/2014 5:02:53 PM. Edward Jackson

Steady stream of bits 9/6/2014 5:02:53 PM. Edward Jackson DB1 As noted in this unit's reading, there are three main categories of streaming video systems. Name the three categories, give a brief comparison of the categories, then state (and support) your own

More information

Capacity Planning for Application Design

Capacity Planning for Application Design WHITE PAPER Capacity Planning for Application Design By Mifan Careem Director - Solutions Architecture, WSO2 1. Introduction The ability to determine or forecast the capacity of a system or set of components,

More information

CS15-319: Cloud Computing. Lecture 3 Course Project and Amazon AWS Majd Sakr and Mohammad Hammoud

CS15-319: Cloud Computing. Lecture 3 Course Project and Amazon AWS Majd Sakr and Mohammad Hammoud CS15-319: Cloud Computing Lecture 3 Course Project and Amazon AWS Majd Sakr and Mohammad Hammoud Lecture Outline Discussion On Course Project Amazon Web Services 2 Course Project Course Project Phase I-A

More information

Effects of Internet Path Selection on Video-QoE

Effects of Internet Path Selection on Video-QoE Effects of Internet Path Selection on Video-QoE by Mukundan Venkataraman & Mainak Chatterjee Dept. of EECS University of Central Florida, Orlando, FL 32826 mukundan@eecs.ucf.edu mainak@eecs.ucf.edu Streaming

More information

QOE ISSUES RELEVANT TO VIDEO STREAMING IN CABLE NETWORKS Jeremy Bennington Praveen Mohandas. Cheetah Technologies, Sunnyvale, CA

QOE ISSUES RELEVANT TO VIDEO STREAMING IN CABLE NETWORKS Jeremy Bennington Praveen Mohandas. Cheetah Technologies, Sunnyvale, CA QOE ISSUES RELEVANT TO VIDEO STREAMING IN CABLE NETWORKS Jeremy Bennington Praveen Mohandas Cheetah Technologies, Sunnyvale, CA Abstract This paper explores how monitoring video quality in a streaming

More information

Performance Analysis of Virtual Machines on NxtGen ECS and Competitive IaaS Offerings An Examination of Web Server and Database Workloads

Performance Analysis of Virtual Machines on NxtGen ECS and Competitive IaaS Offerings An Examination of Web Server and Database Workloads Performance Report: ECS Performance Analysis of Virtual Machines on ECS and Competitive IaaS Offerings An Examination of Web Server and Database Workloads April 215 EXECUTIVE SUMMARY commissioned this

More information

CS 425 / ECE 428 Distributed Systems Fall 2015

CS 425 / ECE 428 Distributed Systems Fall 2015 CS 425 / ECE 428 Distributed Systems Fall 2015 Indranil Gupta (Indy) Measurement Studies Lecture 23 Nov 10, 2015 Reading: See links on website All Slides IG 1 Motivation We design algorithms, implement

More information

Mohammad Hossein Manshaei 1393

Mohammad Hossein Manshaei 1393 Mohammad Hossein Manshaei manshaei@gmail.com 1393 Voice and Video over IP Slides derived from those available on the Web site of the book Computer Networking, by Kurose and Ross, PEARSON 2 multimedia applications:

More information

Qoe-aware adaptive bitrate video streaming over mobile networks with caching proxy

Qoe-aware adaptive bitrate video streaming over mobile networks with caching proxy University of New Brunswick From the SelectedWorks of Jun He March 30, 2015 Qoe-aware adaptive bitrate video streaming over mobile s with caching proxy Jun He, University of New Brunswick Available at:

More information

Microsoft RemoteFX for Remote Desktop Virtualization Host Capacity Planning Guide for Windows Server 2008 R2 Service Pack 1

Microsoft RemoteFX for Remote Desktop Virtualization Host Capacity Planning Guide for Windows Server 2008 R2 Service Pack 1 Microsoft RemoteFX for Remote Desktop Virtualization Host Capacity Planning Guide for Windows Server 2008 R2 Service Pack 1 Microsoft Corporation Published: March 2011 Abstract Microsoft RemoteFX delivers

More information

Streamloading: Low cost high quality video streaming. Shivendra S. Panwar (joint work with Amir Hosseini and Fraida Fund)

Streamloading: Low cost high quality video streaming. Shivendra S. Panwar (joint work with Amir Hosseini and Fraida Fund) Streamloading: Low cost high quality video streaming Shivendra S. Panwar (joint work with Amir Hosseini and Fraida Fund) Modern Cellular Networks Modern Cellular Networks Evolution of modern cellular networks:

More information

HTTP Adap)ve Streaming in prac)ce

HTTP Adap)ve Streaming in prac)ce HTTP Adap)ve Streaming in prac)ce Mark Watson (with thanks to the Ne@lix adapeve streaming team!) ACM MMSys 2011 22-24 February 2011, San Jose, CA NeBlix Overview Started with DVD- by- mail, now primarily

More information

DYNAMIC ADAPTIVE STREAMING OVER HTTP/2.0

DYNAMIC ADAPTIVE STREAMING OVER HTTP/2.0 DYNAMIC ADAPTIVE STREAMING OVER HTTP/2.0 Christopher Mueller, Stefan Lederer, Christian Timmerer, and Hermann Hellwagner Alpen-Adria-Universität Klagenfurt Universitätsstraße 65-67 9020 Klagenfurt am Wörthersee,

More information

HTTP Adaptive Streaming Enhancements for Large-Scale Deployments

HTTP Adaptive Streaming Enhancements for Large-Scale Deployments HTTP Adaptive Streaming Enhancements for Large-Scale Deployments UCLA Nov. 2014 Ali C. Begen For papers and other presentations, visit: http://ali.begen.net Pull-Based Video Delivery over HTTP Progressive

More information

A Converged Content Delivery Platform for IP and QAM Video

A Converged Content Delivery Platform for IP and QAM Video A Converged Delivery Platform for IP and QAM Video Abstract James Barkley, Weidong Mao Comcast Cable HTTP based Adaptive Bit Rate (ABR) video delivery to IP enabled CPEs via Delivery Network (CDN) for

More information

Tail Latency in ZooKeeper and a Simple Reimplementation

Tail Latency in ZooKeeper and a Simple Reimplementation Tail Latency in ZooKeeper and a Simple Reimplementation Michael Graczyk Abstract ZooKeeper [1] is a commonly used service for coordinating distributed applications. ZooKeeper uses leader-based atomic broadcast

More information

Effect of Payload Length Variation and Retransmissions on Multimedia in a WLANs

Effect of Payload Length Variation and Retransmissions on Multimedia in a WLANs Effect of Payload Length Variation and Retransmissions on Multimedia in 8.a WLANs Sayantan Choudhury Dept. of Electrical and Computer Engineering sayantan@ece.ucsb.edu Jerry D. Gibson Dept. of Electrical

More information

WHITEPAPER AMAZON ELB: Your Master Key to a Secure, Cost-Efficient and Scalable Cloud.

WHITEPAPER AMAZON ELB: Your Master Key to a Secure, Cost-Efficient and Scalable Cloud. WHITEPAPER AMAZON ELB: Your Master Key to a Secure, Cost-Efficient and Scalable Cloud www.cloudcheckr.com TABLE OF CONTENTS Overview 3 What Is ELB? 3 How ELB Works 4 Classic Load Balancer 5 Application

More information

A General Sign Bit Error Correction Scheme for Approximate Adders

A General Sign Bit Error Correction Scheme for Approximate Adders A General Sign Bit Error Correction Scheme for Approximate Adders Rui Zhou and Weikang Qian University of Michigan-Shanghai Jiao Tong University Joint Institute Shanghai Jiao Tong University, Shanghai,

More information

Price Performance Analysis of NxtGen Vs. Amazon EC2 and Rackspace Cloud.

Price Performance Analysis of NxtGen Vs. Amazon EC2 and Rackspace Cloud. Price Performance Analysis of Vs. EC2 and Cloud. Performance Report: ECS Performance Analysis of Virtual Machines on ECS and Competitive IaaS Offerings An Examination of Web Server and Database Workloads

More information

Measuring Over-the-Top Video Quality

Measuring Over-the-Top Video Quality Contents Executive Summary... 1 Overview... 2 Progressive Video Primer: The Layers... 2 Adaptive Video Primer: The Layers... 3 Measuring the Stall: A TCP Primer... 4 Conclusion... 5 Questions to Ask of

More information

Mobile Cloud Computing & Adaptive Streaming

Mobile Cloud Computing & Adaptive Streaming Mobile Cloud Computing & Adaptive Streaming 20 th Mar 2012 Suriya Mohan, Aricent Group, Chennai Agenda Mobile Cloud Computing Tablet / Smartphone Evolution Cloud Computing 3 Fundamental Models Clouds in

More information

Network Assisted Content Distribution for Adaptive Bitrate Video Streaming

Network Assisted Content Distribution for Adaptive Bitrate Video Streaming Network Assisted Content Distribution for Adaptive Bitrate Video Streaming Divyashri Bhat, Amr Rizk 2, Michael Zink, Ralf Steinmetz 2 University of Massachusetts Amherst, 2 Technische Universität Darmstadt

More information

A Dynamic Adaptive Algorithm Based on HTTP Streaming Media Technology

A Dynamic Adaptive Algorithm Based on HTTP Streaming Media Technology 2016 International Conference on Electronic Information Technology and Intellectualization (ICEITI 2016) ISBN: 978-1-60595-364-9 A Dynamic Adaptive Algorithm Based on HTTP Streaming Media Technology Zhufeng

More information

Quickly Starting Media Streams Using QUIC

Quickly Starting Media Streams Using QUIC Quickly Starting Media Streams Using QUIC Packet Video Workshop 2018 Şevket Arısu and Ali C. Begen Agenda Motivation and our goal Previous work and our contributions Approach, setup and evaluation Results

More information

WEBSITE & CLOUD PERFORMANCE ANALYSIS. Evaluating Cloud Performance for Web Site Hosting Requirements

WEBSITE & CLOUD PERFORMANCE ANALYSIS. Evaluating Cloud Performance for Web Site Hosting Requirements WEBSITE & CLOUD PERFORMANCE ANALYSIS Evaluating Cloud Performance for Web Site Hosting Requirements WHY LOOK AT PERFORMANCE? There are many options for Web site hosting services, with most vendors seemingly

More information

Delivery of Adaptive Bit Rate Video: Balancing Fairness, Efficiency and Quality

Delivery of Adaptive Bit Rate Video: Balancing Fairness, Efficiency and Quality Delivery of Adaptive Bit Rate Video: Balancing Fairness, Efficiency and Quality Jason J. Quinlan, Ahmed H. Zahran, K.K. Ramakrishnan, Cormac J. Sreenan Dept. of Computer Science, University College Cork,

More information

Octoshape. Commercial hosting not cable to home, founded 2003

Octoshape. Commercial hosting not cable to home, founded 2003 Octoshape Commercial hosting not cable to home, founded 2003 Broadcasting fee is paid by broadcasters Free for consumers Audio and Video, 32kbps to 800kbps Mesh based, bit-torrent like, Content Server

More information

Week-12 (Multimedia Networking)

Week-12 (Multimedia Networking) Computer Networks and Applications COMP 3331/COMP 9331 Week-12 (Multimedia Networking) 1 Multimedia: audio analog audio signal sampled at constant rate telephone: 8,000 samples/sec CD music: 44,100 samples/sec

More information

OpenCache. A Platform for Efficient Video Delivery. Matthew Broadbent. 1 st Year PhD Student

OpenCache. A Platform for Efficient Video Delivery. Matthew Broadbent. 1 st Year PhD Student OpenCache A Platform for Efficient Video Delivery Matthew Broadbent 1 st Year PhD Student Motivation Consumption of video content on the Internet is constantly expanding Video-on-demand is an ever greater

More information

The Guide to Best Practices in PREMIUM ONLINE VIDEO STREAMING

The Guide to Best Practices in PREMIUM ONLINE VIDEO STREAMING AKAMAI.COM The Guide to Best Practices in PREMIUM ONLINE VIDEO STREAMING PART 3: STEPS FOR ENSURING CDN PERFORMANCE MEETS AUDIENCE EXPECTATIONS FOR OTT STREAMING In this third installment of Best Practices

More information

QoS-Aware IPTV Routing Algorithms

QoS-Aware IPTV Routing Algorithms QoS-Aware IPTV Routing Algorithms Patrick McDonagh, Philip Perry, Liam Murphy. School of Computer Science and Informatics, University College Dublin, Belfield, Dublin 4. {patrick.mcdonagh, philip.perry,

More information

A Case Study on Cloud Based Hybrid Adaptive Mobile Streaming: Performance Evaluation

A Case Study on Cloud Based Hybrid Adaptive Mobile Streaming: Performance Evaluation A Case Study on Cloud Based Hybrid Adaptive Mobile Streaming: Performance Evaluation T. Mahesh kumar 1, Dr. k. Santhisree 2, M. Bharat 3, V. Pruthvi Chaithanya Varshu 4 Student member of IEEE, M. tech

More information

Variable Bitrate Stream in Set top Box device

Variable Bitrate Stream in Set top Box device Variable Bitrate Stream in Set top Box device Preeti Chourasia Student M.Tech (CS) United Institute of Technology And Research Greater Noida (UP) Priyank Chourasia MCA (MITS Gwalior) ABSTRACT Video processing

More information

Analysis of WLAN Traffic in the Wild

Analysis of WLAN Traffic in the Wild Analysis of WLAN Traffic in the Wild Caleb Phillips and Suresh Singh Portland State University, 1900 SW 4th Avenue Portland, Oregon, 97201, USA {calebp,singh}@cs.pdx.edu Abstract. In this paper, we analyze

More information

SSD Admission Control for Content Delivery Networks

SSD Admission Control for Content Delivery Networks Technical Disclosure Commons Defensive Publications Series September 20, 2017 SSD Admission Control for Content Delivery Networks Dragos Ionescu Richard Schooler Kenneth Barr Follow this and additional

More information

Performance Characterization of a Commercial Video Streaming Service. Mojgan Ghasemi, Akamai Technologies - Princeton University

Performance Characterization of a Commercial Video Streaming Service. Mojgan Ghasemi, Akamai Technologies - Princeton University Performance Characterization of a Commercial Video Streaming Service Mojgan Ghasemi, Akamai Technologies - Princeton University MGhasemi,PKanuparthy,AMansy,TBenson,andJRexford ACM IMC 2016 1 2 First study

More information

Confused, Timid, and Unstable: Picking a Video Streaming Rate is Hard

Confused, Timid, and Unstable: Picking a Video Streaming Rate is Hard Confused, Timid, and Unstable: Picking a Video Streaming Rate is Hard Araz Jangiaghdam Seminar Networks and Distributed Systems School of Engineering and Sciences Jacobs University Bremen Campus Ring 1,

More information

Using Bandwidth Aggregation to Improve the Performance of Video Quality- Adaptive Streaming Over Multiple Wireless Access Networks

Using Bandwidth Aggregation to Improve the Performance of Video Quality- Adaptive Streaming Over Multiple Wireless Access Networks Using Bandwidth Aggregation to Improve the Performance of Video Quality- Adaptive Streaming Over Multiple Wireless Access Networks 1 J.P. Satheesh, 2 R. Karthikeyan 1 PG Student, 2 Assistant Professor,

More information

Skype Video Responsiveness to Bandwidth Variations

Skype Video Responsiveness to Bandwidth Variations Skype Video Responsiveness to Bandwidth Variations L. De Cicco,, V. Palmisano Dipartimento di Elettronica ed Elettrotecnica Politecnico di Bari Italy -1- Motivation 1/2 Multimedia real-time applications

More information