Networked Life 20 Questions and Answer.pdf-资料库

dmk01101008-7900947-4744300845147083879.pdf-第1页.png

第1页 / 共491页

dmk01101008-7900947-4744300845147083879.pdf-第2页.png

第2页 / 共491页

dmk01101008-7900947-4744300845147083879.pdf-第3页.png

第3页 / 共491页

dmk01101008-7900947-4744300845147083879.pdf-第4页.png

第4页 / 共491页

dmk01101008-7900947-4744300845147083879.pdf-第5页.png

第5页 / 共491页

dmk01101008-7900947-4744300845147083879.pdf-第6页.png

第6页 / 共491页

dmk01101008-7900947-4744300845147083879.pdf-第7页.png

第7页 / 共491页

dmk01101008-7900947-4744300845147083879.pdf-第8页.png

第8页 / 共491页

Networked Life: 20 Questions and Answers Mung Chiang Princeton University April 2012 Draft

Contents Preface Acknowledgements Roadmap What makes CDMA work for my smartphone? How does Google sell ad spaces? How does Google rank webpages? How does Netﬂix recommend movies? When can I trust an average rating on Amazon? Why does Wikipedia even work? How do I viralize a YouTube video and tip a Groupon deal? How do I inﬂuence people on Facebook and Twitter? Can I really reach anyone in 6 steps? Does the Internet have an Achilles’ heel? Why do AT&T and Verizon Wireless charge me $10 a GB? How can I pay less for my Internet connection? How does traﬃc get through the Internet? Why doesn’t the Internet collapse under congestion? How can Skype and BitTorrent be free? 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 page v viii xi 1 25 44 60 88 109 127 155 189 210 230 251 273 306 331

iv Contents 16 17 18 19 20 What’s inside the cloud of iCloud? IPTV and Netﬂix: How can the Internet Support Video? Why is WiFi faster at home than at a hotspot? Why am I only getting a few % of advertised 4G speed? Is it fair that my neighbors iPad downloads faster? Notes Index 354 376 401 427 446 467 469

Preface You pick up your iPhone while waiting in line at a coﬀee shop. You Google a not-so-famous actor and get linked to a Wikipedia entry listing his recent movies and popular YouTube clips. You check out user reviews on IMDB and pick one, download that movie on BitTorrent or stream that in Netﬂix. But suddenly the WiFi logo on your phone is gone and you’re on 4G. Video quality starts to degrade a little, but you don’t know if it’s the video server getting crowded in the cloud or the Internet is congested somewhere. In any case, it costs you $10 per gigabyte, and you decide to stop watching the movie, and instead multitask between sending tweets and calling your friend on Skype, while songs stream from iCloud to your phone. You’re happy with the call quality, but get a little irritated when you see there’re no new followers on Twitter. You’ve got a typical networked life, an online networked life. And you might wonder how all of these technologies “kind of” work, and why sometimes they don’t. Just ﬂip through the table of contents of this book. It’s a mixture: some of these questions have well deﬁned formulations and clear answer while others still face a signiﬁcant gap between the theoretical models and actual practice; a few don’t even have widely-accepted problem statements. This book is about formulating and answering these 20 questions. This book is about the networking technologies we use each day as well as the fundamental ideas in the study of networks. Each question is selected not just for its relevance to our daily lives, but also for the core concepts and key methodologies in the ﬁeld of networking that are illustrated by its answer. These concepts include aggregation and inﬂuence, distributed coordination, feedback control, and strategic equilibrium. And the analytic machineries are based on mathematical languages that people refer to as graph, optimization, game, and learning theories. This is an undergraduate textbook for a new course at Princeton University: Networks: Friends, Money, and Bytes. The course targets primarily juniors in electrical engineering and computer science, but also seniors and beginning graduate students as well as students from mathematics, sciences, economics, and engineering in general. It can be viewed as the second course after the “signals and systems” course that anchors the undergraduate electrical and computer engineering curriculum today. This book weaves a diverse set of topics you would not normally see under

vi Preface the same cover into a coherent stream: from Arrow’s impossibility and Rawls’ fairness to Skype signaling and Clos networks, from collaborative ﬁltering and ﬁreﬂy synchronization to MPEG/RTSP/TCP/IP and WiFi CSMA DCF. This begs a question: “So, what is the discipline of this book?”, a question that most of the undergraduates simply do not care about. Neither does this book: it only wants to address these practical questions, using whatever modeling languages that have been observed to be the most relevant ones so far. Turns out there is a small and coherent set of mathematics we will need, but that’s mostly because people have only invented a limited suite of modeling languages. This is not a typical textbook for another reason. It does not start with general theories as do many books on these subjects, e.g., graph theory, game theory, optimization theory, or abstract concepts like feedback, coordination, and equi- librium. Instead it starts with concrete applications and practical answers, and sticks to them (almost) every step of the way. Theories and generalizations only emerge, as if “accidental by-products”, during the process of formulating and answering these questions. This book, when used as an undergraduate textbook, can be complemented with its website features: http://www.network20q.com, including lecture slides, problem solutions, additional questions, further pointers to references, collec- tion of news media coverage of the topics, currency-earning activities, course projects, blogs, tweets, surveys, and student-generated course materials in wiki. We created web features that turn this class into an online social network and a networked economy. This book can also be used by engineers, technology managers, and pretty much anyone with a keen interest in understanding how social and technologi- cal networks work. On many spots, we sacriﬁce generality for accessibility, and supplement symbolic representation by numerical illustration. • The ﬁrst section of each chapter is a “short answer”, and it is accessible by most people. • Then there’s a “long answer” section. If you remember diﬀerentiation and linear algebra (and occasionally a little bit of integration and basic proba- bility), you can follow all the material there. We take great care to include only those symbols and equations that’re really necessary to unambigiously express the ideas. • The “examples” section contains detailed, numerical examples to reinforce the learning from the “long answer” section. • Each chapter concludes with a section on “advanced material,” which requires the reader to be quite comfortable with symbolic operations and abstract reasoning, but can be skipped without losing the coherence and gist of the book. In the undergraduate course taught at Princeton, almost none of the advanced material is covered. Covering all the advanced material sections would constitute an introductory graduate level course. • At the end of each chapter, there’re 5 homework questions, including easy

Preface vii drills, essential supplements, and some “out-of-syllabus” explorations. The level of diﬃculty is indicated on a scale of 1 (easy) to 3 (hard) stars. • There are also 5 key references per chapter (yes, only 5, in the hope that undergraduates may actually read some of these 5, and my apologies to the authors of thousands of papers and books that could have been cited). These references open the door to many worthwhile further readings, in- cluding textbooks, research monographs, and survey articles. This is a (relatively) thin book. It’s a collage of snapshots, not an encyclopedia. It’s an appetizer, not an entree. We realize that the majority of readers will not pursue a career specializing in the technical material in this book, so we take every opportunity to delete material that’s very interesting to specialists but not essential to this undergraduate course. Each one of these 20 chapters deserves many books for a detailed treatment. We only highlight a few key ideas in the span of about 20 pages per chapter and 80 minutes per lecture. There are many other mathematical languages in the study of networks, many other questions about a networked life, and many other types of networks that we do not have time to cover in one semester. But as the saying goes for a course: “It’s more important to uncover than to cover a lot.” This is a book illustrating some pretty big ideas in networking, through 20 questions we can all relate to in our daily lives. Questions that tickle our imag- ination with surprises and incomplete answers. Questions that I wished I had known how to answer several years ago. Questions that are quickly becoming an essential part of modern education in electrical and computer engineering. But above all, we hope this book is fun to read.

Acknowledgements In so many ways I’ve been enjoying the process of writing this book and creating the new undergraduate course at Princeton University. The best part is that I got to, ironically in light of the content of this book, stay oﬄine and focus on learning a few hours a day for several hundred days. I got to digest wonderful books and papers that I didn’t get a chance to read before, to think about what’re the essential points and simple structures behind the drowning sea of knowledge in my research ﬁelds, and to edit and re-edit each sentence I put down on paper. It reminded me of my own sophomore year, one and half decade ago, at Stanford University. I often biked to the surreally beautiful Oval in the morning and dived into books of many kinds, most of which not even remotely related to my majors. As the saying goes, that was a pretty good approximation of paradise. That paradise usually ends together with the college years. So I have many to thank for granting me a precious opportunity to indulge myself again at this much later stage in life. • The new course “Networks: Friends, Money, and Bytes” could not have been created without the dedication from its three remarkable TAs: Jiasi Chen, Felix Wong, and Pei-yuan Wu. They did so much more for the course than a “normal” TA experience. • Many students and postdocs in Princeton’s EDGE Lab and EE Department worked with me in creating worked examples: Chris Brinton, Amitabha Ghosh, Sangtae Ha, Joe Jiang, Carlee Joe-Wong, Yiannis Kamitsos, Haris Kremo, Chris Leberknight, Soumya Sen, Arvid Wang, and Michael Wang. • Princeton students in ELE/COS 381’s ﬁrst oﬀering were brave enough to take a completely new course and contributed in many ways, not the least the class website blogs and course projects. Students in the graduate course ELE539A also helped proofread the book draft and created multiple choice questions. • Before I even get a chance to advertise the course, some colleagues started planning to oﬀer similar courses at their institutions: Jianwei Huang (CUHK, Hong Kong), Hongseok Kim (Sogang U., Korea), Tian Lan (GWU), Walid Saad (U. Miami), Chee Wei Tan (City U., Hong Kong), Kevin Tang (Cor- nell), more... • Over 50 colleagues provided valuable suggestions to the course and the book.

资料库

Networked Life 20 Questions and Answer.pdf

相关推荐

课程资源

热门标签

最新资料