This is a book about Ethernet, a local area network (LAN) technology that allows you to connect a variety of computers together with a low-cost and extremely flexible network system. Virtually every computer manufacturer today supports Ethernet, and this broad support, coupled with its low cost and high flexibility, are major reasons for Ethernet’s popularity.
This book provides a comprehensive and practical source of information on the entire Ethernet system in a single volume. The goal of this book is to be definitive: to describe the entire range of Ethernet technology specified in the IEEE standard for Ethernet. This includes 10 Mbps Ethernet, 100 Mbps Fast Ethernet, 1000 Mbps Gigabit Ethernet, full-duplex Ethernet, descriptions of all Ethernet media systems, and repeaters and repeater configuration guidelines. Also described in this book are switching hubs, structured cabling systems, network management, troubleshooting and more.
This book shows how Ethernet components can be combined to create Ethernet LANs. While some basic network designs are shown in this book, there are an infinity of network designs that can be built using Ethernet, ranging from the smallest workgroup on up to very large enterprise networks that support tens of thousands of computers.
The design of complete network systems that use Ethernet to carry data between computers is a major subject, and a number of books are needed to describe all of the issues that can be encountered. Since this book is about how Ethernet technology works, we stay focused on that topic. As anyone who reads the entire book would agree, this topic alone has more than enough detail for any single book to cover.
The Ethernet system has grown over the years, becoming ever larger and more complex. It now includes a wide variety of media systems, each based on its own particular set of hardware and each with its own configuration guidelines. This book covers all Ethernet systems that have ever been widely implemented, from the latest Gigabit Ethernet systems all the way back to the original coaxial cable systems. With this book you can support the entire range of Ethernet technology you may encounter.
As the Ethernet system has grown more complex, a number of misconceptions and misunderstandings have arisen about how Ethernet functions and how the system should be configured. To provide the most accurate information possible and to help combat incorrect “Ethernet folklore,” I kept a complete set of official Ethernet standards at my elbow while writing this book and referred to them frequently. I have been working with Ethernet technology since the early 1980s, and that experience has included many hard-won lessons in network design and operation that have also made their way into this book.
There are a number of factors that have helped Ethernet to become so popular. Among these factors are cost, scalability, reliability, and widely available management tools.
The rapid evolution of new capabilities in Ethernet has also been accompanied by a rapid decrease in the cost of Ethernet equipment. The widespread adoption of Ethernet technology created a large and fiercely competitive Ethernet marketplace, which drives down the cost of networking components. As a result, the consumer wins out in the process, with the marketplace providing a wide range of competitively priced Ethernet components to choose from.
The first industry-wide Ethernet standard was published in 1980. This standard defined a 10 Mbps system, which was very fast for the times, and which remained fast enough for most uses until the mid-1990s. The development of the 100 Mbps Fast Ethernet system in 1995 provided a tenfold increase in speed. Fast Ethernet has been a major success, and network interfaces that can automatically support both 10 and 100 Mbps operation are widely available, making the transition from 10 Mbps to 100 Mbps systems very easy to accomplish.
Applications tend to grow to fill all available bandwidth. To anticipate the rising demand, Gigabit Ethernet was developed in 1998, providing yet another tenfold increase in performance. All of this makes it possible for a network manager to provide high-speed backbone systems and connections to high-performance servers. Desktop machines can be connected to the original 10 Mbps Ethernet, 100 Mbps Fast Ethernet, or Gigabit Ethernet as required.
Ethernet uses a simple and robust transmission mechanism that reliably delivers data day in and day out at sites all over the world. Ethernet based on twisted-pair media was introduced in 1987, making it possible to provide Ethernet signals over a structured cabling system. Structured cabling provides a data delivery system for a building that is modeled on high-reliability cabling practices originally developed for the telephone system. This makes it possible to install a standards-based cabling system for Ethernet that is very reliable, as well as being simple, stable, and easy to manage.
The widespread acceptance of Ethernet brings another advantage, which is the wide availability of Ethernet management and troubleshooting tools. Management tools based on standards, such as the Simple Network Management Protocol (SNMP), make it possible for network administrators to keep track of an entire campus full of Ethernet equipment from a central location. Management capabilities embedded in Ethernet repeaters, switching hubs, and computer interfaces provide powerful network monitoring and troubleshooting capabilities.
A major goal of this book is to help you design and implement reliable networks. Network reliability is of paramount importance to any networked organization. Information sharing between networked computers is an essential feature of today’s workplace, and if the network fails, everything comes to a halt. This book shows you how to design reliable networks, how to monitor them and keep them working reliably, and how to fix them should something fail.
The wide range of Ethernet components and cabling systems that are available today provides enormous flexibility, making it possible to build an Ethernet to fit just about any circumstance. However, all this flexibility does have a price. The many varieties of Ethernet each have their own components and their own configuration rules, which can make the life of a network designer complex. Designing and implementing a reliable Ethernet system requires that you understand how all the bits and pieces fit together, and that you follow the official guidelines for the configuration of the media systems.
This book provides the complete set of official configuration guidelines for every commercially available media system, as well as the official guidelines for combining media systems. You’ll also find a great deal of information on how to build media systems that meet the standards and that will function reliably.
Avoiding network downtime is important for a number of reasons, not least of which is the amount of money that downtime can cost. Some quick “back of the envelope” calculations can show how expensive network downtime can be. Let’s assume that there are 1,000 users of the network at the Amalgamated Widget Company, and that their average annual salary including all overhead (benefits, pension, etc.) is $75,000. That comes to $75 million a year in employee costs.
Let’s further assume that everyone in the company depends on the network to get their work done, and that the network is used 40 hours a week, for about 50 weeks of the year (excluding holidays). That’s 2,000 hours of network operation. Dividing the annual employee cost by the hours of network operation shows that the network is supporting $37,500 per hour of employee cost during the year.
When we total up all of the network outages over the period of a year in our hypothetical corporation, we find that the network was down 2.5 percent of the time. That’s an annual total of 50 hours, or one hour a week, or a mere 12 minutes each day. Fifty hours of network downtime at $37,500/hour is $1.8 million in lost productivity due to network outage. Obviously, our calculations are very “quick and dirty.” We didn’t bother to calculate the impact of network outages during times when no one is around, but during which times the network is still supporting critically important servers. Also, we’re assuming that a network failure brings all operations to a halt, instead of trying to factor in the varying effects of localized failures that cause outages on only a portion of the network system. Nor do we try to estimate how much other work people could get done while the network is down, which would tend to lessen the impact.
However, the main point is clear: even small amounts of network downtime can cost a company quite a lot in lost productivity. That’s why it’s worth investing extra time, effort and money to create the most reliable network system you can afford.
The purpose of this book is to provide a comprehensive and practical guide to the entire Ethernet system. The emphasis is on practical issues, with minimal theory and jargon. Chapters are kept as self-contained as possible, and many examples and illustrations are provided. The book is organized in five parts to make it easier to find the specific information you need.
These five parts provide:
An introduction to the Ethernet standard which describes Ethernet operation in detail. This part of the book covers those portions of Ethernet operation that are common to all Ethernet media systems.
A description of each of the Ethernet media systems, including 10-, 100-, and 1000 Mbps systems operating over twisted-pair and fiber optic cables. The older coaxial media systems are described in Appendix B.
A description of structured cabling systems and the components and cables used in building your Ethernet system. These include twisted-pair and fiber optic cables, and repeater and switching hubs.
A description of Ethernet performance and Ethernet troubleshooting.
Appendixes and glossary.
Chapter 1 through Chapter 5 provide a tour of basic Ethernet theory and operation. This section includes the portions of Ethernet operation that are common to all of the Ethernet media systems, including the structure of the Ethernet frame and the operation of the media access control (MAC) system.
- Chapter 1
Gives a brief guide to the history of Ethernet and the development of the IEEE 802.3 standard for Ethernet.
- Chapter 2
Presents an overview of how the Ethernet system operates, introducing the major concepts.
- Chapter 3
Provides an in-depth look at how the original half-duplex Ethernet channel operates.
- Chapter 4
Describes the full-duplex mode of Ethernet operation.
- Chapter 5
Describes the auto-negotiation mechanisms used to automatically configure Ethernet equipment.
Chapter 6 provides an introduction to the basic media system components used in all Ethernet media systems. This chapter is essential reading before going on to the individual media systems, described in Chapter 7 through Chapter 12. Chapter 13 completes this part of the book with a description of the configuration guidelines that apply when linking media systems together with repeaters.
Each of the media system chapters are based on an identical format, which helps to organize and clearly present the information needed to cover all of the Ethernet media varieties. While every effort was made to avoid needless duplication of text, the identical format leads to some unavoidable repetition in these chapters. This is especially noticeable if you read several media chapters in a row.
- Chapter 6
Describes the Ethernet media components and the basic concepts that are common to each of the media systems.
- Chapter 13
Describes the official guidelines for combining media varieties using repeaters.
Chapter 14 describes the structured cabling standards. Chapter 15 and Chapter 16 provide information on the configuration and construction of twisted-pair and fiber optic cable segments; Chapter 17 and Chapter 18 describe how to design and build Ethernet systems using repeaters and switches.
- Chapter 14
Describes structured cabling systems and the structured cabling standards that specify how these systems are built.
- Chapter 15
Describes the twisted-pair cables and components used in twisted-pair network segments.
- Chapter 16
Describes the fiber optic cables and components used in fiber optic network segments.
- Chapter 17
Describes the operation and management of Ethernet repeater hubs and how to design networks using them.
- Chapter 18
Describes the operation and management of Ethernet switching hubs and how to design networks using them.
- Appendix A
Describes additional sources of information on Ethernet, including books, periodicals, and web sites.
- Appendix B
Describes the thick and thin coaxial media systems and hardware components.
- Appendix C
Describes equipment and configuration issues based on the original 15-pin Ethernet AUI.
Provides concise definitions of the acronyms and technical terms relevant to Ethernet.
A number of online references are provided in this book, based on the Universal Resource Locators (URLs) used on the World Wide Web. Web references are live in the sense that the Web is constantly evolving and changing, which may render a reference obsolete. Sometimes a replacement link will be left, pointing to the new location for the information. If that happens, all you have to do is click on the new link to find what you’re looking for.
Other times a site may be reorganized in a manner that leaves no forwarding link to the new location. If an online reference no longer works, you can try several approaches to finding the material.
One method is to access the top-level web page by using the first part of the URL, which specifies the domain name of the site. For example, if the URL http://www.bellereti.com/ethernet/ethernet.html should fail to work, you could try just the domain name portion of the URL, located inside the first set of slashes, http://www.bellereti.com/, and see what you find there.
Some web sites may also be equipped with a search feature that allows you to type in the name of the material you are looking for at that site. If all else fails, you can try one of the many web search sites that will search the entire Web for the subjects you’re looking for.
The goal of this book is to provide the information needed for you to understand and operate any Ethernet system. For example, if you are a newcomer to Ethernet and you need to know how twisted-pair Ethernet systems work, then you can start with the chapters in Part I. After reading those chapters, you can go to the twisted-pair media chapters in Part II, as well as the twisted-pair cabling information in Part III. Twisted-pair segments can be connected together with both repeater hubs and switching hubs, and these are also described in Part III. Experts in Ethernet can use the book as a reference guide and jump directly to those chapters that contain the reference information they need.
Constant Willisonis used for program examples, attribute value literals, start- and end-tags, and source code example text.
Constant Willison Obliqueis used for “replaceable” text or variables. Replacement text is text that describes something you’re supposed to type, like a
filename, in which the word “filename” acts as a placeholder for the actual filename.
Garamond Italic is used for filenames and URLs.
URLs (http://www.oreilly.com/) are presented in parenthesis after the name of the resource they describe in the book.
We have tested and verified the information in this book to the best of our ability, but you may find that features have changed (or even that we have made mistakes!). Please let us know about any errors you find, as well as your suggestions for future editions, by writing to:
|O’Reilly & Associates, Inc.|
|101 Morris Street|
|Sebastopol, CA 95472|
|1-800-998-9938 (in the U.S. or Canada)|
You can also send us messages electronically. To be put on the mailing list or request a catalog, send email to:
To ask technical questions or comment on the book, send email to:
The web site for Ethernet: The Definitive Guide lists errata and plans for future editions. You can access this page at:
For more information about this book and others, see the main O’Reilly web site:
This book would not have been possible without the help of many people. First and foremost, I would like to thank the inventors of Ethernet, Bob Metcalfe, and his fellow researchers at Xerox PARC. Their work revolutionized the way computers are used, unleashing a powerful new communications technology based on information sharing applications running on computers linked with networks. I also thank the many engineers who have voluntarily given their time in countless IEEE standards meetings to help develop the Ethernet system and to write the Ethernet specifications.
I would like to thank Mark Stone, executive editor for O’Reilly’s Open Source editorial group, for his interest in this project and for all the work that he and his colleagues at O’Reilly & Associates have put into making this book possible. Chuck Toporek at O’Reilly has spent many hours applying his copyediting skills to excellent effect, for which I thank him. Thank you to Hanna Dyer for the cover design, and David Futato, the production editor for this book. Chris North, Eric Pearce, Jesse Robbins, and Rich Seifert provided reviews of the manuscript that helped improve the final work. It’s difficult for busy people to find time to provide a detailed review and to compile useful responses for a large manuscript such as this one, and I am especially grateful to the reviewers who were able to do so. Of course, I alone am responsible for any remaining errors.
Finally, I wish to thank my wife, Joann Zimmerman, for enduring yet another book project, and for her patience, her unstinting support, and her editing skills. Without her very able assistance, this book would not have been possible.