Cover Page

ffirsuf001

Advanced Content Delivery, Streaming, and Cloud Services

 

 

Edited by

 

Mukaddim Pathan

 

Telstra Corporation Ltd., Australia

 

Ramesh K. Sitaraman

 

University of Massachusetts, Amherst and Akamai Technologies, USA

 

Dom Robinson

 

id3as-company Ltd., UK

 

 

 

 

 

 

Wiley Logo

To my wife Ziyuan for her inspiration, love, and support. This book would not have been completed, if she did not single-handedly take care of everything, while I was too busy in writing and compilation!—Mukaddim

To my wife Vidya and our active children Anu and Siddu for whom the Internet can never be fast enough!—Ramesh

To my wife Mariana and our wonderful kids Sofia and Zac—I am sure you will enjoy this book as a bedtime reading. And to my parents (that funny “computer thing” you bought me as a kid came in handy!)—Dom

Preface

The ever-evolving nature of the Internet brings new challenges in managing and delivering content to end-users. Content Delivery Networks (CDNs) improve Web access and streaming performance, in terms of response time and system throughput, while delivering content to Internet end-users through multiple, geographically distributed edge servers. The CDN industry, that is, content delivery, consumption, and monetization, has been undergoing rapid changes. The multidimensional surge in content delivery from end-users has led to an explosion of new content, formats, and an exponential increase in the size and complexity of the digital content supply chain. These changes have been accelerated by economic downturn in that the content providers are under increasing pressure to reduce costs while increasing revenue.

The main value proposition for CDN services has shifted over time. Initially, the focus was on improving end-user-perceived experience by decreasing response time, especially when the customer website experiences unexpected traffic surges. Nowadays, CDN services are treated by content providers as a way to use a shared infrastructure to handle their peak capacity requirements, thus allowing reduced investment cost in their own hosting infrastructure. Moreover, recent trends in CDNs indicate a large paradigm shift toward a utility computing model, which allows customers to exploit advanced content delivery services, hosted on commodity hardware, without having to build a dedicated infrastructure.

From a market perspective, historically buyers based the bulk of demand and spending on “core” CDN products that facilitate the delivery of Web-based content services. Over the last few years, offering from video streaming and value-added services (VASs) peaked as the most demandable CDN products. They formed the basis of most of the present-day CDNs' offering, while strong demand for the basic CDN services still continues. Market research shows that on average, buyers reported 43% of total CDN spending on core products, such as caching and content delivery, while 57% spending on VAS-based products.

While satisfying the market demands, CDN providers are more and more focusing on higher margin, VAS offering in order to gain (or stabilize) overall profit margins. These VASs include mobile data acceleration, content protection, content management, application acceleration, mobile data delivery, and cloud-based storage. While these products currently have a reasonable market penetration, they represent even more substantial near-term growth opportunities.

In addition to the emergence of innovative CDN models, such as managed CDNs, licensed CDNs, and federated CDNs, Telco/operator CDNs are evolving into major market share holders. Telcos/operators around the world have started building CDN platform, technology, and support to aid content consumption, delivery, and rich media experience by end-users. The geographic expansion of Web-based content continues to grow and drive global CDN business requirements. Many CDN players have started with a regional focus and then expanded to offer services in new regions. It is expected that these trends in the CDN industry will continue, as the definition and scope of a CDN gets broader.

1.1 Overview and Scope of the Book

The book entitled Advanced Content Delivery, Streaming, and Cloud Services presents fundamental and trendy CDN technologies, with a comprehensive coverage of evolution, current landscape, and future roadmap. The book builds on academic and industrial research and developments, and case studies that are being carried out at different organizations around the world. In addition, the book identifies potential research directions and technologies that will drive future innovations. This book is aimed at a large audience including systems architects, practitioners, product developers, and researchers. It can be used as a reference/textbook for graduate students and a roadmap for academicians, who are starting to research in the field of content delivery. We expect the readers to have at least the basic knowledge about Web technologies and the Internet. In particular, readers should be knowledgeable about Web caching, replication, Internet-based services and applications, and basic networking.

Upon reading this, book readers will perceive the following benefits:

  1. Learn the state of the art in research and development on content management, delivery, and streaming technologies.
  2. Obtain a future roadmap by learning open research issues.
  3. Gather the background knowledge to tackle key problems, whose solutions will enhance the evolution of next-generation content networks.
  4. Use the book as a valuable reference and/or textbook.

1.2 Organization of the Book

This book is organized into three parts, namely, Part I: CDN and Media Streaming Basics; Part II: CDN Performance Management and Optimization; and Part III: Case Studies and Next-Generation CDNs. Specifically, the topics of the book are the following:

Part I of the book focuses on the basic ideas, techniques, and current practices related to content delivery and media streaming. Chapter 1 by Pathan presents an overview of CDNs, operational models, and use cases. It covers recent market and technology trends, as well as identifies a few research issues in the CDN domain. Robinson, in Chapters 2 and 3, provides a comprehensive description of the live media streaming ecosystem and demonstrates the practical configuration of live streaming using a few tools. In Chapter 4, Haßlinger identifies key properties of caching and content delivery in broadband access network, and describes how efficiency can be achieved by configuration and performance tuning. Alzoubi et al. in Chapter 5 present mechanisms and algorithms to effectively redirect end-user requests in a CDN platform. This chapter demonstrates the applicability of IP anycasting for request redirection. Basics of content delivery to cloud-based home ecosystem is covered in Chapter 6 by Cruz et al., highlighting key challenges, industry practices, and recent trends. In Chapter 7, Narayanan et al. describe the challenges in delivering video in mobile networks and present various adaptation techniques for mobile video streaming.

Part II of the book provides a coverage of CDN performance measurement techniques, tools, reporting, and analytics. In Chapter 8, Siglin covers CDN analytics tools and explores a variety of analytic practices and their implications in practical context, including new methods for analyzing adaptive bitrate (ABR) streaming technology. Mathematical modeling to optimize CDN services, such as video on demand (VoD) content delivery, is covered in Chapter 9 by Bektaimage and Ercetin. It makes the reader aware of fundamental optimization problems arising in content delivery and the ways of effectively solving these problems. Molina et al. in Chapter 10 present a basic analytical model to analyze the basic and advanced properties of a CDN. Zhanikeev in Chapter 11 describes a method for cloud-based multisource streaming and compares its performance over traditional methods. In Chapter 12, Islam and Grégoire discuss on the intersection of CDN and cloud computing by exposing a number of trade-offs on the deployment of multimedia processing functions inside the cloud and identify relevant performance factors. In Chapter 13, Yoshida describes the performance of a dynamic streaming CDN, comprising techniques for dynamic network reorganization, and load distribution and balancing to realize dynamicity, as well as techniques for stream segmentation and reconstruction, and QoS assurance. Cesario et al. in Chapter 14 present the analysis of mining streaming data in a CDN, improving efficiency and effectiveness of a CDN architecture. A hybrid multidomain architecture is described that solves the problem of computing frequent items and frequent itemsets from distributed data streams. In Chapter 15, Davies and Pathan cover the capacity planning process that is instrumental for the ongoing operation of a deployed CDN infrastructure. It includes a practical application and workflow of the CDN capacity planning process.

Part III, the final part of the book, consists of a handful of representative case studies on present- and next-generation CDNs. In Chapter 16, Sitaraman et al. discuss different network overlays that are crucial for meeting the needs for Internet-based services. Architecture and techniques of representative overlays are discussed, along with their practical usage and implications. Chapter 17 by Pai provides coverage of a variety of next-generation CDNs and presents a case study of CoBlitz, a research CDN that became a commercial licensed CDN. In Chapter 18, Talyansky et al. describe the challenges of content delivery in China, by drawing on experience from ChinaCache, a carrier-neutral CDN. A brief coverage of content-aware network services offered by ChinaCache is provided, along with future trends of content delivery within China. Chapter 19, by Czyrnek et al., presents a case study of a high definition (HD) interactive TV platform, called PlatonTV. In addition to describing the PlatonTV architecture, different aspects of content delivery such as content ingest, content distribution, and management within the CDN are discussed. In Chapter 20, Srebrny et al. present CacheCast—a link layer caching system for single-source, multiple destination data transfers. In this case study, CacheCast architecture, operational methodology, and deployment details are presented. Sourlas et al. in Chapter 21 present a generic three-phase framework for content replication in information centric networks (ICNs). Algorithms supporting efficient replication in ICN are discussed and performance benefits are demonstrated. Chapter 22 by Fortino et al. describes content delivery techniques in vehicular ad hoc networks (VANets). A content broadcasting methodology is presented, which improves content transfer time and delivery efficiency in the radio network. Finally, in Chapter 23, Kilanioti et al. discuss approaches to leverage information from online social networks (OSNs) for rich media content delivery in CDNs. Future research directions in this area, along with a few commercial implications for CDNs, are also discussed.

Mukaddim Pathan

Telstra Corporation Ltd., Australia

Ramesh K. Sitaraman

University of Massachusetts, Amherst and Akamai Technologies, USA

Dom Robinson

id3as-company Ltd., UK

Acknowledgments

This book came into light because of the direct and indirect involvement of many researchers, academics, and industry practitioners. We acknowledge and thank the contributing authors; research institutions; and companies whose papers, reports, articles, notes, websites, and study materials have been referred to in this book. We are thankful to Professor Albert Zomaya, editor of the Wiley Series on Parallel and Distributed Computing, for his support in accepting the book proposal and guiding us through Wiley's publication process. We express our gratitude to Simone Taylor, Director, Editorial Development of John Wiley & Sons, Inc., and Wiley's publication staff, for handling the book project and ensuring a timely publication.

All chapters were peer reviewed, and authors have updated their chapters addressing the review comments. Prior technical sources are acknowledged citing them at appropriate places in the book. In case of any errors, we would like to receive feedback so that it could be taken into consideration in the next edition.

We hope that this book will serve as a valuable text for students especially at graduate level and a reference for researchers and practitioners working in the content delivery domain.

Mukaddim, Ramesh, and Dom

Contributors

Hussein A. Alzoubi Case Western Reserve University, Cleveland, OH, USA

Tolga Bektaimage University of Southampton, Highfield, Southampton, UK

Carlos T. Calafate Universitat Politècnica de València, Valencia, Spain

Jaime Calvo Universidad de Salamanca, Escuela Politecnica Superior de Zamora, Zamora, Spain

Juan C. Cano Universitat Politècnica de València, Valencia, Spain

Eugenio Cesario ICAR-CNR, Rende (CS), Italy

Tiago Cruz Faculdade de Ciências e Tecnologia da, Universidade de Coimbra, Coimbra, Portugal

Mirosław Czyrnek Poznan Supercomputing and Networking Center, Poznaimage, Poland

Phil Davie Telstra Corporation Limited, Melbourne, Victoria, Australia

Ozgur Ercetin Sabancı University, imagestanbul, Turkey

Manuel Esteve Universitat Politecnica de Valencia, Valencia, Spain

Paris Flegkas University of Thessaly, Oktovriou, Volos, Greece

Giancarlo Fortino University of Calabria, Rende (CS), Italy

Chryssis Georgiou Department of Computer Science, University of Cyprus, Nicosia, Cyprus

Vera Goebel University of Oslo, Oslo, Norway

Jean-Charles Grégoire INRS-EMT, Montréal, QC, Canada

Gerhard Haßlinger Deutsche Telekom Technik, Darmstadt, Germany

Salekul Islam United International University, Dhaka, Bangladesh

Manish Jain Akamai Technologies, Inc., Cambridge, MA, USA

Jimagedrzej Jajor Poznan Supercomputing and Networking Center, Poznaimage, Poland

Jerzy Jamroimagey Poznan Supercomputing and Networking Center, Poznaimage, Poland

Mangesh Kasbekar Akamai Technologies, Inc., Cambridge, MA, USA

Dimitrios Katsaros University of Thessaly, Oktovriou, Volos, Greece

Anuj Kaul Nokia Siemens Networks, Mountain View, CA, USA

Irene Kilanioti Department of Computer Science, University of Cyprus, Nicosia, Cyprus

Ewa Kuimagemierek Poznan Supercomputing and Networking Center, Poznaimage, Poland

Seungjoon Lee AT&T Labs—Research, Florham Park, NJ, USA

Woody Lichtenstein Akamai Technologies, Inc., Cambridge, MA, USA

Pietro Manzoni Universitat Politècnica de València, Valencia, Spain

Carlo Mastroianni ICAR-CNR, Rende (CS), Italy

Andreas Mauthe InfoLab 21, Lancaster University, Lancaster, UK

Cezary Mazurek Poznan Supercomputing and Networking Center, Poznaimage, Poland

Benjamin Molina Universitat Politecnica de Valencia, Valencia, Spain

Edmundo Monteiro Faculdade de Ciências e Tecnologia da, Universidade de Coimbra, Coimbra, Portugal

Ram Lakshmi Narayanan Nokia Siemens Networks, Mountain View, CA, USA

Vivek S. Pai Princeton University, Princeton, NJ, USA

Carlos E. Palau Universitat Politecnica de Valencia, Valencia, Spain

George Pallis Department of Computer Science, University of Cyprus, Nicosia, Cyprus

Mukaddim Pathan Telstra Corporation Ltd., Melbourne, Victoria, Australia

Thomas Plagemann University of Oslo, Oslo, Norway

Michael Rabinovich Case Western Reserve University, Cleveland, OH, USA

Dom Robinson id3as-company Ltd., Rottingdean, Brighton, Sussex, UK

Mili Shah Nokia Siemens Networks, Mountain View, CA, USA

Timothy Siglin Braintrust Digital, Inc., Harriman, TN, USA

Paulo Simões Faculdade de Ciências e Tecnologia da, Universidade de Coimbra, Coimbra, Portugal

Ramesh K. Sitaraman University of Massachusetts, Amherst, and Akamai Technologies, MA, USA

Dag H.L. Sørbø University of Oslo, Oslo, Norway

Vasilis Sourlas University of Thessaly, Oktovriou, Volos, Greece

Oliver Spatscheck AT&T Labs—Research, Florham Park, NJ, USA

Piotr Srebrny University of Oslo, Oslo, Norway

Maciej Stroiimageski Poznan Supercomputing and Networking Center, Poznaimage, Poland

Domenico Talia ICAR-CNR, Rende (CS), Italy; DIMES, University of Calabria, Rende (CS), Italy

Michael Talyansky ChinaCache, Sunnyvale, CA, USA

Leandros Tassiulas University of Thessaly, Oktovriou, Volos, Greece

Alexei Tumarkin ChinaCache, Sunnyvale, CA, USA

Kobus Van Der Merwe University of Utah, Salt Lake City, UT, USA

Jan Wimageglarz Poznan Supercomputing and Networking Center, Poznaimage, Poland

Hunter Xu ChinaCache, Beijing, China

Yinghua Ye Nokia Siemens Networks, Mountain View, CA, USA

Norihiko Yoshida Information Technology Center, Saitama University, Saitama, Japan

Ken Zhang ChinaCache, Beijing, China

Marat Zhanikeev Kyushu Institute of Technology, Iizuka, Japan

Part I

CDN and Media Streaming Basics