Scalable Network Monitoring in High Speed Networks

ISBN-13: 9781489985637 / Angielski / Miękka / 2014 / 148 str.

Baek-Young Choi;Zhi-Li Zhang;David Hung-Chang Du

Opis

Network monitoring serves as the basis for a wide scope of network, engineering and management operations. Precise network monitoring involves inspecting every packet traversing in a network. However, this is not feasible with future high-speed networks, due to significant overheads of processing, storing, and transferring measured data. Network Monitoring in High Speed Networks presents accurate measurement schemes from both traffic and performance perspectives, and introduces adaptive sampling techniques for various granularities of traffic measurement. The techniques allow monitoring systems to control the accuracy of estimations, and adapt sampling probability dynamically according to traffic conditions. The issues surrounding network delays for practical performance monitoring are discussed in the second part of this book. Case studies based on real operational network traces are provided throughout this book. Network Monitoring in High Speed Networks is designed as a secondary text or reference book for advanced-level students and researchers concentrating on computer science and electrical engineering. Professionals working within the networking industry will also find this book useful.

Recenzje

From the reviews:

"In great detail, the authors discuss the notions of flow and of elephants versus mice in the context of network traffic. ... The authors discuss the shortcomings of static sampling and packet counting, and note the important contribution the packet size has on the load. ... Altogether, the insights covered in the book should be interesting to network designers." (J. S. Edwards, ACM Computing Reviews, December, 2011)

Spis treści

Overview.- Part I Scalable Traffic Monitoring with Packet Sampling.- Load Characterization and Measurement.- Flow Characterization and Measurement.- Part II Scalable Delay Monitoring with Active Probing.- Analysis of Point-To-Point Delay in an Operational Backbone.