What do you think?
Rate this book
Computer Networking: A Top-down Approach Featuring the Internet & Distributed Systems, Principles & Paradigms
Certain data-communication protocols hog the spotlight, but they all have a lot in common. Computer Networking: A Top-Down Approach Featuring the Internet explains the engineering problems inherent in communicating digital information from point to point. The top-down approach mentioned in the subtitle means the book starts at the top of the Open Systems Interconnection (OSI) protocol stack--with the application layer--and works its way down through the other six layers until it reaches bare wire. The approach is definitely theoretical--don't look here for instructions on configuring Windows 2000 or a Cisco router--but it is relevant to reality and should help anyone who needs to understand networking as a programmer, system architect or even administration guru.
The treatment of the network layer, where routing takes place, is typical of the style overall. In discussing routing, authors Kurose and Ross explain (by way of lots of clear, definition-packed text) what routing protocols need to do: find the best route to a destination. They then present the mathematics that determine the best path, show some C code that implements those algorithms and illustrate the logic with excellent conceptual diagrams. Real-life implementations of the algorithms--including Internet Protocol (both IPv4 and IPv6) and several popular IP routing protocols--help you make the transition from pure theory to networking technologies. --David Wall
Paperback
First published November 1, 2000
485 people are currently reading
4050 people want to read
Ratings & Reviews
Friends & Following
Create a free account to discover what your friends think of this book!
Community Reviews
Displaying 1 - 30 of 136 reviews
August 22, 2014
This textbook was required reading for a Computer Networking course. As far as college textbooks go, it was pretty good. I learned the material from it that I needed to learn. The textbook was well-organized, and it was easy to understand. It had quite a few analogies and real-world examples that aided in understanding.
For me, this textbook was a pretty dry read and I sometimes had trouble pushing through the weekly reading assignments. However, networks are not an aspect of Computer Science that I’m particularly interested in, so this was to be expected. I usually found the beginning of each chapter interesting, but my interest tapered off about halfway through and the rest was a chore to get through. However, I did learn a lot about how networks work. I may not retain all of the nitty-gritty details over the long term, but I expect the main concepts to stick with me.
I think one of the best parts of this textbook were the Wireshark labs at the companion website, which my professor used for some of our assignments. I had fun learning how to “sniff packets”, and I liked seeing the protocols in action for myself using real, live information being passed to and from my own personal network.
Aside from my subjective complaint about the dryness of the material, my only other real complaint is in regard to the acronyms. There were an amazing number of acronyms, and many of them were not in the index so it could be difficult to find the original definitions again. Once an acronym was defined once, it was not defined again – at least not within the chapter in which it was introduced. (The chapters were 70-100 pages long.) A glossary of acronyms in the back of the book would have really helped. To any future readers of this textbook, I recommend making a note of each acronym you encounter and at least noting the page # where it was first introduced in case you need to reference its definition again.
For me, this textbook was a pretty dry read and I sometimes had trouble pushing through the weekly reading assignments. However, networks are not an aspect of Computer Science that I’m particularly interested in, so this was to be expected. I usually found the beginning of each chapter interesting, but my interest tapered off about halfway through and the rest was a chore to get through. However, I did learn a lot about how networks work. I may not retain all of the nitty-gritty details over the long term, but I expect the main concepts to stick with me.
I think one of the best parts of this textbook were the Wireshark labs at the companion website, which my professor used for some of our assignments. I had fun learning how to “sniff packets”, and I liked seeing the protocols in action for myself using real, live information being passed to and from my own personal network.
Aside from my subjective complaint about the dryness of the material, my only other real complaint is in regard to the acronyms. There were an amazing number of acronyms, and many of them were not in the index so it could be difficult to find the original definitions again. Once an acronym was defined once, it was not defined again – at least not within the chapter in which it was introduced. (The chapters were 70-100 pages long.) A glossary of acronyms in the back of the book would have really helped. To any future readers of this textbook, I recommend making a note of each acronym you encounter and at least noting the page # where it was first introduced in case you need to reference its definition again.
February 27, 2021
This was such a boring book. So much so that I would discourage anyone from picking ip up (or studying Computer Networking for that matter). First let me point out the characteristics that I liked most, so that I won't come across as a nuisance, and that prevented me from giving a lower rating
- At anyone's reach
- Very comprehensive prose, as far as the topics that the authors chose to cover are concerned (indeed, I felt much was left over)
Now comes my critique: this book feels aimed at a five-year old audience! Paragraphs feel like a watered-down, wordy mass of text -- this gives clarity, but is boring as hell to read! I expected (if not demanded from a Computer Science course) more formalism, technical sophistication than what I got from two authors afraid of overwhelming their audience. Computer networking basics are covered satisfactorily, but the watered-down, romanticized prose prevented the text from achieving a thicker conceptual density; a book style more appropriate for students with an interest in industry-related applications than on insight and true understanding. I once had a talk with a networking tutor at my university, and he told me part of his work dealt with Markov's chains and probabilistic modeling; I read many of the end-of-chapter interviews, and the interviewees often remarked a conceptually richer footing for computer networking that this book expressed. Why wasn't any of all this covered in this text?
Where did the cool stuff go? OK, I got a good grasp on the ISO/OSI stack -- up from HTTP, e-mail systems, P2P networks down to TCP congestion and flow control mechanisms, link-layer devices and LANs and basic network security notions (that was probably the part I liked most) but in the end I felt I was only acquiring passive knowledge with no reasoning involved. Is that a problem with this particular book? With computer networking? Worse yet, for computer science courses in general?
Additional flaws
- Ridiculously over-priced hard-cover copy (I borrowed one from the library)
- No solutions for end-of-chapter exercises (not even wireshark)
- Questionably ugly top-down approach. Makes one feel like applications deserve more attention than theoretical understanding
- At anyone's reach
- Very comprehensive prose, as far as the topics that the authors chose to cover are concerned (indeed, I felt much was left over)
Now comes my critique: this book feels aimed at a five-year old audience! Paragraphs feel like a watered-down, wordy mass of text -- this gives clarity, but is boring as hell to read! I expected (if not demanded from a Computer Science course) more formalism, technical sophistication than what I got from two authors afraid of overwhelming their audience. Computer networking basics are covered satisfactorily, but the watered-down, romanticized prose prevented the text from achieving a thicker conceptual density; a book style more appropriate for students with an interest in industry-related applications than on insight and true understanding. I once had a talk with a networking tutor at my university, and he told me part of his work dealt with Markov's chains and probabilistic modeling; I read many of the end-of-chapter interviews, and the interviewees often remarked a conceptually richer footing for computer networking that this book expressed. Why wasn't any of all this covered in this text?
Where did the cool stuff go? OK, I got a good grasp on the ISO/OSI stack -- up from HTTP, e-mail systems, P2P networks down to TCP congestion and flow control mechanisms, link-layer devices and LANs and basic network security notions (that was probably the part I liked most) but in the end I felt I was only acquiring passive knowledge with no reasoning involved. Is that a problem with this particular book? With computer networking? Worse yet, for computer science courses in general?
Additional flaws
- Ridiculously over-priced hard-cover copy (I borrowed one from the library)
- No solutions for end-of-chapter exercises (not even wireshark)
- Questionably ugly top-down approach. Makes one feel like applications deserve more attention than theoretical understanding
April 23, 2023
I've picked up this book because I wanted to improve my knowledge about computer networks.
I think that the book is very approachable, the authors are using very good analogies in some places to make the material easier to understand. The main part of the book tries to cover the Application Layer, Transport layer, Network layer and Link layer. Some things(e.g. protocols) in the book are explained very well, while others are just mentioned. I find this normal because computer networking is an interdisciplinary field and the authors had to pick which topics to focus on if they wanted to write a <1000 pages book (The references section in the book is 40 pages).
In the end I learnt a lot from this book(I could have learnt even more if I were to do all of the exercises at the end of each chapter) and even though it took me a lot of time to read I am glad that I did it.
I think that the book is very approachable, the authors are using very good analogies in some places to make the material easier to understand. The main part of the book tries to cover the Application Layer, Transport layer, Network layer and Link layer. Some things(e.g. protocols) in the book are explained very well, while others are just mentioned. I find this normal because computer networking is an interdisciplinary field and the authors had to pick which topics to focus on if they wanted to write a <1000 pages book (The references section in the book is 40 pages).
In the end I learnt a lot from this book(I could have learnt even more if I were to do all of the exercises at the end of each chapter) and even though it took me a lot of time to read I am glad that I did it.
April 29, 2013
Nice and understandable. Makes you excited about the Internet. I enjoy reading the interviews featured in every chapter (like the one with Marc Andreessen included in the sixth edition).
December 13, 2024
literally read cover to cover for my exam. great explanations and examples, wish I could’ve read it at a more leisure pace
November 20, 2022
Great book to learn about internet, a marvelous invention. It took more than 20 years for this layer 0 to find the killer use case - the Web.
For the layer 1, we are still early.
For the layer 1, we are still early.
August 8, 2023
The book is pretty long and verbose (but digestible / accessible), but remember, you have access to his ~2022 YouTube videos here:
https://www.youtube.com/@JimKurose
Some nice YouTube people even curated some pretty organized custom Playlists.
Some textbook sections are missing from the video series, but there is this channel that can fill in the gaps - also a PhD lecturer: https://www.youtube.com/@EpicNetworks
He explains everything brilliantly, the textbook will supplement details if certain concepts elude you from the video. I think this is one of the authoritative textbooks on the topic. I enjoyed it. Very affable instructor - very passionate, knowledgeable, and upbeat about the topic.
Thanks!
https://www.youtube.com/@JimKurose
Some nice YouTube people even curated some pretty organized custom Playlists.
Some textbook sections are missing from the video series, but there is this channel that can fill in the gaps - also a PhD lecturer: https://www.youtube.com/@EpicNetworks
He explains everything brilliantly, the textbook will supplement details if certain concepts elude you from the video. I think this is one of the authoritative textbooks on the topic. I enjoyed it. Very affable instructor - very passionate, knowledgeable, and upbeat about the topic.
Thanks!
December 17, 2018
(Re 7th edition)
The authors successfully make the deep details of networking understandable and even easy to read.
Love the “a day in the life of a webpage request”, which listed the 24 exact detailed steps that a packet takes between an ethernet card, a Web server and it’s returned. (p501-505)
Chapter 8 on security leaves a lot to be desired… It’s really about how security is designed to ideally function, but omits all the real world problems. Similarly the last chapters on mobile and multimedia feel incomplete and rushed.
The authors successfully make the deep details of networking understandable and even easy to read.
Love the “a day in the life of a webpage request”, which listed the 24 exact detailed steps that a packet takes between an ethernet card, a Web server and it’s returned. (p501-505)
Chapter 8 on security leaves a lot to be desired… It’s really about how security is designed to ideally function, but omits all the real world problems. Similarly the last chapters on mobile and multimedia feel incomplete and rushed.
October 18, 2019
If you want to learn how computer networks work, I think this is a wonderful introduction! I enjoyed every single page of the book. and the book is not just a great textbook about computer networks, but also a great example of how teaching should be! You could always feel the tutors by your side, you could feel they have been in your place and understand your confusion and misunderstandings. So if you need a book to learn how to teach, I think this is a great example of it too.
March 10, 2019
Read some chapters from this book for my networking course. The writers explained the nuts and bolts of networking in such interesting way, I felt like I was reading a story! Thanks to them for explaining the hard topics so easily that we, students clung to the textbook..
December 20, 2024
Heel droog en zwaar om allemaal front-to-back te lezen, maar wel heel duidelijk. Een goeie intro + deep dive in wat networking eigenlijk is en welke protocollen er allemaal bestaan. Er zou wel een overzichtstabel mogen bestaan van alle belangrijke begrippen (afkortingen etc...) neem dus zeker notities als je dit leest
March 11, 2025
This book has been following me around for months like a looming specter, except the terror is just 700 pages of highly technical writing. Amazing book, don't get me wrong! It provides just about as much color as one can to the topic of computer networking, but god do I feel freed now.
March 19, 2025
The best networks book! So good. So much content almost too much but the information was super useful and the pictures were great! Thank you kurose!
February 19, 2025
An amazing book that teaches networking top down as the name suggests. Starts with Application Layer and walks you down all the way to the PHY layer. Talks about TCP in and out and covers concepts like MSS, Fragmentation at L3. LAN, VLANS and how L2 came into existence, from hubs having CSMA/CD to Switches where there are no collisions. Covers WiFi later in the later part of the book explaining beautifully the concepts of SSID, BSSID and signal strengths. Later, it dwelves into security teaches basics of cryptography.
An amazing and a must read book for anyone who wants to get into networking. A must read.
One of those book that you read, reread and read again to reread.
An amazing and a must read book for anyone who wants to get into networking. A must read.
One of those book that you read, reread and read again to reread.
This entire review has been hidden because of spoilers.
November 28, 2025
Leí este libro por un curso universitario de redes de computadoras. Sus capítulos pueden ser extensos para un libro académico, pero tantas secciones hacen que se vuelva rápido de leer... aunque yo terminara atrasado en las lecturas. El libro se centra en el modelo TCP/IP, stack de protocolos de Internet basado en 5 capas: aplicación, transporte, red, enlace y física.
Leí en alguna review sugerencias sobre incluir un glosario de acrónimos, y en verdad es cierto que tomando apuntes puede ser útil anotarlos en un lugar conocido para evitar tener que —como yo— volver atrás cuando aparecen acrónimos poco referenciados porque, ¿qué era DOCSIS?
El capítulo 1 introduce muchos conceptos que, aunque interesantes, son desconocidos para quienes apenas se adentran en las redes de computadoras y puede resultar estresante. El valor de este capítulo lo encontramos a medida seguimos la lectura, y nos deja la intuición de muchos conceptos importantes.
El capítulo 2, sobre la capa de aplicación desarrolla los conceptos de la capa de aplicación e introduce la capa de transporte, cosa que se vuelve recurrente en Kurose dado que al desarrollar la teoría de una capa también da intuiciones de la capa inferior, manteniéndose fiel a la arquitectura en capas del modelo TCP/IP. Es un libro plagado de acrónimos y de analogías pero facilitan la comprensión. Leerlo con presión no es ideal, pero aún así se hace ameno.
Los problemas al final del capítulo son muy útiles, y se presentan muchas herramientas para diagnóstico de redes. Algunos de los protocolos tratados son: HTTP, SMTP con POP3 e IMAP, P2P (BitTorrent, particularmente) y DNS. Kurose aprovecha a introducir la noción de sockets y ejercicios variados, por ejemplo sobre la noción de proxys que también se desarrolla en el capítulo, así como la de cookies.
El capítulo 3 puede llegar a ser de los más densos en la teoría pero de los más interesantes. Kurose hace el desarrollo de un protocolo de capa de transporte de transferencia de datos confiable genérico denominado rdt (reliable data transfer) el cual desarrolla incrementalmente en sus versiones rdt 1.0, 2.0, 2.1, 2.2 y termina con una versión rdt 3.0 "similar" a TCP. Kurose comienza por distinguir la comunicación entre procesos en un mismo sistema terminal ya sea mediante memoria compartida o IPC (Inter Process Communication) de la comunicación entre distintos hosts/sistemas terminales/end systems. Desarrolla brevemente sobre UDP (User Datagram Protocol) como un protocolo muy simple de capa de transporte cuya función es la mínima esperada de capa de transporte, la de establecer una conexión lógica proceso-proceso (para cada proceso es transparente cómo se da la comunicación). El desarrollo de TCP (Transport Control Protocol) es más extenso, y se centra principalmente en la transferencia confiable de datos, el control de flujo y el control de congestión. En este punto es donde Kurose deja muchos temas abiertos, aunque los menciona.
Kurose no se adentra en qué hacer con el tráfico UDP en Internet que podría expulsar el tráfico TCP, pero frecuentemente referencia a otros autores. También hay que considerar que la edición que he leído es de hace varios años ya (aunque al menos esta versión dejó de usar a Skype como ejemplo para todo). El control de congestión de TCP se explica de forma detenida pero noté muy apresurada la distinción entre las versiones de Tahoe y Reno; considero que dos AFD (Deterministic Finite Automaton, DFA) hubieran esclarecido esto. Como es costumbre, se agrega la noción de capa de red muy temprano en este capítulo, lo que nos prepara para lo que sigue, y las analogías no se detienen.
El capítulo 4 introduce la noción de capa de red, centrándose en el plano de datos, en la tarea de reenvío (forwarding) de los routers y la tecnología dentro de ellos. También se distingue IPv4 de IPv6, pero la teoría expuesta sobre IPv6 me pareció insuficiente. En particular, las ideas de tunneling no se profundizan, pero en alguna referencia se trata 6in4, 6to4, Tunneling Browser y tunneling de Teredo de forma algo pobre, aunque puede que se deba a mi inexperiencia. Fuera de esto, el capítulo se hace muy ágil.
También se tratan conceptos de ICMPv6 (y se retoma ICMP en el capítulo 5), DHCP y NAT, entre otros tantos, que son de inmenso interés y me fueron de mucha utilidad para revisar algunos ajustes de mi router doméstico. ¡WOW, APLICANDO REDES!
El capítulo 5 retoma desde el capítulo 4 la capa de red centrado en el plano de control, en los routing algorithms y routing protocols. Se desarrollan LS (Link State) y DV (Distance Vector) como algoritmos, OSPF (Open Shortest Path First) y BGP (Border Gateway Protocol) como protocolos que ejemplifican ambos algoritmos, respectivamente de LS y DV. La descripción de los mensajes intercambiados en ambos protocolos fue nula en la edición que leí (la 7th) y desconozco cómo es en el resto, pero es algo de esperar de un libro que toca tantos temas y, en verdad, solo su superficie. En mi edición solo leí IP Anycast, pero en ediciones más recientes se incluye multicast y broadcast, que también leí y estuvo muy bien.
Kurose desarrolla las ideas de broadcast con inundación no controlada que genera un "broadcast storm"; la inundación controlada mediante dos estrategias: mediante números de secuencia y mediante RPF (Reverse Path Forwarding); y broadcast con árboles recubridores, y una estrategia de construcción center-based tree (determinar un nodo centro y construir el árbol a partir de rutas que lo tienen como destino) usando mensajes tipo join desde los demás routers, cuyas rutas seguidas son las que se añaden al árbol. Multicast usa ideas similares: un árbol recubridor compartido sin distinguir entre emisores, o distinguiendo entre emisores volviendo a usar RPF pero añadiendo pruning (no se contemplan nodos ni ramas que no tienen miembros de un grupo multicast, lo que determina mediante el uso del protocolo Internet Group Management Protocol, IGMP, entre routers y hosts) para evitar propagar datagramas que no llegarían a miembros del grupo multicast particular.
El capítulo 6 trata temas muy interesantes, en particular los de VLAN aunque brevemente. Ignoré algunos aspectos como MPLS pero sin dudas un capítulo muy bien desarrollado. Aún así, considero que en el protocolo de acceso múltiple de toma de turnos mediante polling Kurose podría ahondar un poco más respecto a adueñarse del canal, y tratar alguna implementación particular. Puede que sea un error conceptual de mi parte, pero me dio la impresión de que esto es posible dado que Kurose señala «The master node can determine when a node has finished sending its frames by observing the lack of a signal on the channel», de modo que si uno no respeta el máximo de tramas podría “monopolizar” el canal. La idea de usar timers me pareció acertada pero viendo implementaciones en general se suele asumir que algo así no ocurre, así que terminé suponiendo que serán usados en redes bien administradas y por ello Kurose lo omitió... pero una aclaración me habría ayudado.
Sobre los capítulos 7, 8 y 9 leí algo pero vagamente. Sin dudas algo que me queda pendiente es el capítulo 8 sobre seguridad y del capítulo 7 algo muy particular sobre las IP móviles, porque la idea de establecer conexiones TCP mientras un dispositivo se desplaza entre redes aparenta ser un dolor de cabeza.
Conclusión: Es un muy buen libro de introducción, considero que su idea de aproximación descendente se cumple perfectamente. Para aclarar dudas conviene tener material adicional y revisar documentos RFC para protocolos “abiertos”. Las tareas de laboratorio vi que se limitan a Wireshark, pero recomiendo practicar con Mininet de Stanford buscando la tarea de laboratorio de implementar algún routing protocol, ARP y el manejo de paquetes IP en una red emulada.
No son 5★ por los inconvenientes que mencioné, pero si la idea es leerlo como introducción a las redes y se pueden permitir los vacíos teóricos en algunas secciones, las 5★ le quedan perfectas.
Leí en alguna review sugerencias sobre incluir un glosario de acrónimos, y en verdad es cierto que tomando apuntes puede ser útil anotarlos en un lugar conocido para evitar tener que —como yo— volver atrás cuando aparecen acrónimos poco referenciados porque, ¿qué era DOCSIS?
El capítulo 1 introduce muchos conceptos que, aunque interesantes, son desconocidos para quienes apenas se adentran en las redes de computadoras y puede resultar estresante. El valor de este capítulo lo encontramos a medida seguimos la lectura, y nos deja la intuición de muchos conceptos importantes.
El capítulo 2, sobre la capa de aplicación desarrolla los conceptos de la capa de aplicación e introduce la capa de transporte, cosa que se vuelve recurrente en Kurose dado que al desarrollar la teoría de una capa también da intuiciones de la capa inferior, manteniéndose fiel a la arquitectura en capas del modelo TCP/IP. Es un libro plagado de acrónimos y de analogías pero facilitan la comprensión. Leerlo con presión no es ideal, pero aún así se hace ameno.
Los problemas al final del capítulo son muy útiles, y se presentan muchas herramientas para diagnóstico de redes. Algunos de los protocolos tratados son: HTTP, SMTP con POP3 e IMAP, P2P (BitTorrent, particularmente) y DNS. Kurose aprovecha a introducir la noción de sockets y ejercicios variados, por ejemplo sobre la noción de proxys que también se desarrolla en el capítulo, así como la de cookies.
El capítulo 3 puede llegar a ser de los más densos en la teoría pero de los más interesantes. Kurose hace el desarrollo de un protocolo de capa de transporte de transferencia de datos confiable genérico denominado rdt (reliable data transfer) el cual desarrolla incrementalmente en sus versiones rdt 1.0, 2.0, 2.1, 2.2 y termina con una versión rdt 3.0 "similar" a TCP. Kurose comienza por distinguir la comunicación entre procesos en un mismo sistema terminal ya sea mediante memoria compartida o IPC (Inter Process Communication) de la comunicación entre distintos hosts/sistemas terminales/end systems. Desarrolla brevemente sobre UDP (User Datagram Protocol) como un protocolo muy simple de capa de transporte cuya función es la mínima esperada de capa de transporte, la de establecer una conexión lógica proceso-proceso (para cada proceso es transparente cómo se da la comunicación). El desarrollo de TCP (Transport Control Protocol) es más extenso, y se centra principalmente en la transferencia confiable de datos, el control de flujo y el control de congestión. En este punto es donde Kurose deja muchos temas abiertos, aunque los menciona.
Kurose no se adentra en qué hacer con el tráfico UDP en Internet que podría expulsar el tráfico TCP, pero frecuentemente referencia a otros autores. También hay que considerar que la edición que he leído es de hace varios años ya (aunque al menos esta versión dejó de usar a Skype como ejemplo para todo). El control de congestión de TCP se explica de forma detenida pero noté muy apresurada la distinción entre las versiones de Tahoe y Reno; considero que dos AFD (Deterministic Finite Automaton, DFA) hubieran esclarecido esto. Como es costumbre, se agrega la noción de capa de red muy temprano en este capítulo, lo que nos prepara para lo que sigue, y las analogías no se detienen.
El capítulo 4 introduce la noción de capa de red, centrándose en el plano de datos, en la tarea de reenvío (forwarding) de los routers y la tecnología dentro de ellos. También se distingue IPv4 de IPv6, pero la teoría expuesta sobre IPv6 me pareció insuficiente. En particular, las ideas de tunneling no se profundizan, pero en alguna referencia se trata 6in4, 6to4, Tunneling Browser y tunneling de Teredo de forma algo pobre, aunque puede que se deba a mi inexperiencia. Fuera de esto, el capítulo se hace muy ágil.
También se tratan conceptos de ICMPv6 (y se retoma ICMP en el capítulo 5), DHCP y NAT, entre otros tantos, que son de inmenso interés y me fueron de mucha utilidad para revisar algunos ajustes de mi router doméstico. ¡WOW, APLICANDO REDES!
El capítulo 5 retoma desde el capítulo 4 la capa de red centrado en el plano de control, en los routing algorithms y routing protocols. Se desarrollan LS (Link State) y DV (Distance Vector) como algoritmos, OSPF (Open Shortest Path First) y BGP (Border Gateway Protocol) como protocolos que ejemplifican ambos algoritmos, respectivamente de LS y DV. La descripción de los mensajes intercambiados en ambos protocolos fue nula en la edición que leí (la 7th) y desconozco cómo es en el resto, pero es algo de esperar de un libro que toca tantos temas y, en verdad, solo su superficie. En mi edición solo leí IP Anycast, pero en ediciones más recientes se incluye multicast y broadcast, que también leí y estuvo muy bien.
Kurose desarrolla las ideas de broadcast con inundación no controlada que genera un "broadcast storm"; la inundación controlada mediante dos estrategias: mediante números de secuencia y mediante RPF (Reverse Path Forwarding); y broadcast con árboles recubridores, y una estrategia de construcción center-based tree (determinar un nodo centro y construir el árbol a partir de rutas que lo tienen como destino) usando mensajes tipo join desde los demás routers, cuyas rutas seguidas son las que se añaden al árbol. Multicast usa ideas similares: un árbol recubridor compartido sin distinguir entre emisores, o distinguiendo entre emisores volviendo a usar RPF pero añadiendo pruning (no se contemplan nodos ni ramas que no tienen miembros de un grupo multicast, lo que determina mediante el uso del protocolo Internet Group Management Protocol, IGMP, entre routers y hosts) para evitar propagar datagramas que no llegarían a miembros del grupo multicast particular.
El capítulo 6 trata temas muy interesantes, en particular los de VLAN aunque brevemente. Ignoré algunos aspectos como MPLS pero sin dudas un capítulo muy bien desarrollado. Aún así, considero que en el protocolo de acceso múltiple de toma de turnos mediante polling Kurose podría ahondar un poco más respecto a adueñarse del canal, y tratar alguna implementación particular. Puede que sea un error conceptual de mi parte, pero me dio la impresión de que esto es posible dado que Kurose señala «The master node can determine when a node has finished sending its frames by observing the lack of a signal on the channel», de modo que si uno no respeta el máximo de tramas podría “monopolizar” el canal. La idea de usar timers me pareció acertada pero viendo implementaciones en general se suele asumir que algo así no ocurre, así que terminé suponiendo que serán usados en redes bien administradas y por ello Kurose lo omitió... pero una aclaración me habría ayudado.
Sobre los capítulos 7, 8 y 9 leí algo pero vagamente. Sin dudas algo que me queda pendiente es el capítulo 8 sobre seguridad y del capítulo 7 algo muy particular sobre las IP móviles, porque la idea de establecer conexiones TCP mientras un dispositivo se desplaza entre redes aparenta ser un dolor de cabeza.
Conclusión: Es un muy buen libro de introducción, considero que su idea de aproximación descendente se cumple perfectamente. Para aclarar dudas conviene tener material adicional y revisar documentos RFC para protocolos “abiertos”. Las tareas de laboratorio vi que se limitan a Wireshark, pero recomiendo practicar con Mininet de Stanford buscando la tarea de laboratorio de implementar algún routing protocol, ARP y el manejo de paquetes IP en una red emulada.
No son 5★ por los inconvenientes que mencioné, pero si la idea es leerlo como introducción a las redes y se pueden permitir los vacíos teóricos en algunas secciones, las 5★ le quedan perfectas.
March 18, 2025
A phenomenal read for anyone looking to learn the ins and outs of computer networking. As a student tying to kill some time, I picked up this book to challenge the little I knew in the way of networking, but having finished it, I can say I hardly ever found its contents challenging at all (In a good way). Although comprehensive and detailed, the book builds upon simple ideas accompanied by everyday examples to create an intuitive understanding of modern computer networking in both its academic and applied settings. The book’s questions further reinforce both theory and practice, with test-your-knowledge style questions accompanied by some cool wireshark labs which have the added benefit of making you feel like the le epic hacker you always dreamed of. Some questions and references, as expected with old-ish textbooks, lead to now defunct links, but for the most part, not a big deal.
Additionally the book is meticulous, very meticulous, and I would say overly so except perhaps to the most dedicated of networking enthusiasts. In fairness, there’s only so much someone can do to make memorizing packet headers fun, but for the average person like myself I would say that certain sections of the book are best served as a good reference rather than as a thorough read. Don’t get me wrong though, the book’s delivery is, for the most part, surprisingly funny and witty. It even shames me to admit to say I’ve laughed at a couple of the jokes they sprinkle throughout the text which I really didn’t expect going into this.
Overall, for the average undergraduate student, this book is a great walkthrough/resource for anyone looking to get a real grasp on the why’s and how’s of the internet, and I would strongly recommend it for this purpose.
Additionally the book is meticulous, very meticulous, and I would say overly so except perhaps to the most dedicated of networking enthusiasts. In fairness, there’s only so much someone can do to make memorizing packet headers fun, but for the average person like myself I would say that certain sections of the book are best served as a good reference rather than as a thorough read. Don’t get me wrong though, the book’s delivery is, for the most part, surprisingly funny and witty. It even shames me to admit to say I’ve laughed at a couple of the jokes they sprinkle throughout the text which I really didn’t expect going into this.
Overall, for the average undergraduate student, this book is a great walkthrough/resource for anyone looking to get a real grasp on the why’s and how’s of the internet, and I would strongly recommend it for this purpose.
May 3, 2013
Great form of an introduction to the networking field, perhaps a little too simplified for an IT student, but overtly simplistic enough for someone embarking on the journey down the bit-stream. This might be the books strength, since the harder topics are easily described and easier do not need any real-life translations. The book follows the idea of top-down approach through both ranges of topics discussed and the language used, as the abstraction level is further abstracted to real-life terminology anyone can grasp fast and easily. The lower we get, the deeper in to the secrets of the network and the more technical stuff is discussed. Still, the information is easy to follow since one already must have easily understood the previous topics. Yet if you're into more technical stuff or network management geeky specification stuff this just isn't for you. If you're only a beginner, these almost 780 insight pages may allow you to understand the ideas and cope with most of the tasks network-related. After all, this book makes for a basis of a university/college course.
February 12, 2019
Wow, it was very long reading. It took me about a year to read it.
There are a lot of information about the network today and it is very
difficult to grab all main ideas from the various domains.
I would say that I like first 5 chapters. They cover all important details
from the application layer till the Ethernet frames. After 5 fundamental
chapters authors offer us more domain-specific information. These chapters are
very basic and require a lot of additional reading.
As a result if your goal is to grab the main fundamental concepts you will
be OK with first 5 chapters.
There are a lot of information about the network today and it is very
difficult to grab all main ideas from the various domains.
I would say that I like first 5 chapters. They cover all important details
from the application layer till the Ethernet frames. After 5 fundamental
chapters authors offer us more domain-specific information. These chapters are
very basic and require a lot of additional reading.
As a result if your goal is to grab the main fundamental concepts you will
be OK with first 5 chapters.
August 11, 2020
A very clear and thorough overview of computer networking. Covers the OSI model of networking, as well as the protocols used in each layer. After reading this, I feel that I understand how all modern devices communicate and share data with each other. Very easy to read, even for a beginner, but also extremely thorough. The problems at the end of each chapter were also very good, and helped cement the understanding gained from the text.
August 23, 2007
Only an idiot would read this book.
November 29, 2018
Best. Book. Ever.
June 14, 2025
Stupid Book.
The book contains a large number of the author's subjective arguments and omits important content, while the first chapter devotes extensive space to unimportant details.
It is also riddled with errors and filled with lengthy, baseless discussions that are completely unfounded.
For example, in Chapter 1, it says that the propagation time of a frame is unrelated to the number of links—that is, it's independent of the distance between the endpoints. According to the algorithm in the book, the transmission rate of the circuit is calculated as the number of frames transmitted per second multiplied by the number of bits per timeslot in each frame. However, the book doesn't specify whether "the number of frames transmitted per second" refers to the number of frames sent per second or the number of frames in transit between the two endpoints per second. If it refers to the number of frames sent per second, then indeed it has nothing to do with distance, since that’s determined when the connection is established between the two endpoints. But if propagation time is said to be independent of distance, does that mean the frame reaches the other endpoint instantly upon being sent, without being affected by the physical distance at all? That’s the only way I can interpret it.
This book also says: "When one end system has data to send to another end system, the sending end system segments the data and adds header bytes to each segment. The resulting packages of information, known as packets in the jargon of computer networks, are then sent through the network to the destination end system, where they are reassembled into the original data."
This statement is not accurate — bytes are not only added at the header. If the author were being more precise, they should have referred to it as a data segment, since a data segment can refer to the final data format after bytes have been added, before entering the Internet.
The book contains a large number of the author's subjective arguments and omits important content, while the first chapter devotes extensive space to unimportant details.
It is also riddled with errors and filled with lengthy, baseless discussions that are completely unfounded.
For example, in Chapter 1, it says that the propagation time of a frame is unrelated to the number of links—that is, it's independent of the distance between the endpoints. According to the algorithm in the book, the transmission rate of the circuit is calculated as the number of frames transmitted per second multiplied by the number of bits per timeslot in each frame. However, the book doesn't specify whether "the number of frames transmitted per second" refers to the number of frames sent per second or the number of frames in transit between the two endpoints per second. If it refers to the number of frames sent per second, then indeed it has nothing to do with distance, since that’s determined when the connection is established between the two endpoints. But if propagation time is said to be independent of distance, does that mean the frame reaches the other endpoint instantly upon being sent, without being affected by the physical distance at all? That’s the only way I can interpret it.
This book also says: "When one end system has data to send to another end system, the sending end system segments the data and adds header bytes to each segment. The resulting packages of information, known as packets in the jargon of computer networks, are then sent through the network to the destination end system, where they are reassembled into the original data."
This statement is not accurate — bytes are not only added at the header. If the author were being more precise, they should have referred to it as a data segment, since a data segment can refer to the final data format after bytes have been added, before entering the Internet.
January 30, 2025
Really good and easy read about the way computer networks operate from the Top-Down. It explains it in a way that's both relatable and easy to understand. Found it a bit dry in some parts due to how long each chapter is and my main gripe is the lack of more practical examples. My main field isn't networks so I'm not sure if Network Engineers deal with this level of theoretical stuff, such as connecting the router and letting it do the heavy lifting.
I feel a bit disappointed because I expected lots of coding exercises and/or circuit immersion to fully understand network protocols from the fundamentals but maybe the book leans more towards a general audience with no background that wants to get a good idea of what networks are. I enjoyed the discussion about network protocols and particularly the cryptography exercises but wish we saw more interactive ways to build the protocol or modify the data.
Overall good experience, now to try with more advanced books to see the protocols on a more technical scale.
I feel a bit disappointed because I expected lots of coding exercises and/or circuit immersion to fully understand network protocols from the fundamentals but maybe the book leans more towards a general audience with no background that wants to get a good idea of what networks are. I enjoyed the discussion about network protocols and particularly the cryptography exercises but wish we saw more interactive ways to build the protocol or modify the data.
Overall good experience, now to try with more advanced books to see the protocols on a more technical scale.
March 19, 2018
A better one would have been the bottom up approach. In the beginning of the book it starts to explain the concept of routing and the interconnection of the networks. If a reader has no prior knowledge about networks then he will be scratching his head. In my case i did not even know the actual difference between the router and the switch, where does routing actually takes place, on which layer, which addressing is used. The book directly jumps into the application layer without even giving the introduction of the different layers, brief introduction of them, data forms such as frame, datagram, segment on different layers.
Read
August 22, 2020It's an interesting book about Computer Networking. in my opinion starting from the Application layer backward to the lowest level layers (presentation, Session, etc.) is an impressive and useful approach.
Maybe if you begin from the lowest layers. You will be lost with many boring details!
In General, Computer Science is full of details and maybe boring for many people, but getting into it with a semantic approach will make it more interesting and give it meaning.
I am not against the bottom-up approach, but it must be driven by practice and examples, without doing that it will be boring and time-wasting!
Maybe if you begin from the lowest layers. You will be lost with many boring details!
In General, Computer Science is full of details and maybe boring for many people, but getting into it with a semantic approach will make it more interesting and give it meaning.
I am not against the bottom-up approach, but it must be driven by practice and examples, without doing that it will be boring and time-wasting!
August 26, 2020
The first time I read parts of this book was during my undergraduate studies. I decided to re-read after almost 10 years (a later version of it) and I didn't regret it. Even though I had extensive hands-on experience with computer networking since the first read, I still got to learn new things from the book and gain new insights. The only reason my rating is a 4-star and not a 5-star is because I felt the book tried to cover some areas (e.g. physical layer, multimedia & mobile networking) very superficially for the sake of completeness leaving the reader confused and needing to do more reading to understand what the book was talking about.
January 18, 2018
نمیتونم بگم کتاب بدی بود. برای کسی که دانش پایه و درک و شناخت کلی از شبکه داره خوندن این کتاب کمک زیادی به فهم عمیقتر و جزییترش میکنه. اما برای کسی که از صفر داره شروع میکنه ممکنه اذیتکننده و زننده باشه.
بزرگترین مشکل کتاب زیاد بودن تکرار مکرراتشه. مفهومی رو که ابتدای صفحه توضیح داده، انتهای صفحه دوباره «کامل» توضیح میده و برای خواننده ملال آورش میکنه.
البته لازمه بگم تمرینات برنامهنویسی و وایرشارک انتهای هر فصل خالی از لطف نیستن.
بزرگترین مشکل کتاب زیاد بودن تکرار مکرراتشه. مفهومی رو که ابتدای صفحه توضیح داده، انتهای صفحه دوباره «کامل» توضیح میده و برای خواننده ملال آورش میکنه.
البته لازمه بگم تمرینات برنامهنویسی و وایرشارک انتهای هر فصل خالی از لطف نیستن.
October 30, 2020
Starting from the application layer definitely makes it a great introduction for software-focused students. The textbook is easy to read and understand, although sometimes unnecessarily verbose. The online student resources (especially the programming assignments) are very useful as they bridge the gap between understanding the core concepts from the textbook and being able to read the RFCs properly.
July 10, 2021
A thorough, updated treatment of the technical details of computer networking. If considered strictly as a textbook, rather than a book about technology, I'd probably give it five stars. I'm giving it four stars because it is written rather dryly. Still, it's probably the most comprehensive and reliable book I can think of on the subject, and it's been exceedingly helpful to me over the years in its various editions for coming to grips with network design and topology.
October 5, 2021
Generally thorough and takes a conversational tone throughout most of the text, making for an easier read. Highlights included coverage of contemporary networking technologies (as of 2020), and a fairly solid introduction to key cryptographic primitives as a bonus in the 8th chapter on network security; pitfalls included some spotty editing in places and coverage of mobile networks that felt a bit too glossy despite getting its own chapter. Solid textbook for the material overall, though.
Displaying 1 - 30 of 136 reviews