Analysis of TCP Performance in Data Center Networks

ISBN-13: 9781461478607 / Angielski / Miękka / 2013 / 87 str.

Santosh Kulkarni; Prathima Agrawal

Opis

This book addresses the need to improve TCP's performance inside data centers by providing solutions that are both practical and backward compatible with standard TCP versions. The authors approach this challenge first by deriving an analytical model for TCP's performance under typical data center workload traffic. They then discuss some solutions that are designed to improve TCP performance by either proactively detecting network congestion through probabilistic retransmission or by avoiding timeout penalty through dynamic resizing of TCP segments. Experimental results show that each of techniques discussed outperforms standard TCP inside a data center.

Recenzje

From the reviews:
"This book precisely discusses the performance of TCP in data center networks and methods to overcome the TCP incast problem. In general, the concepts, problems, and solutions related to the performance of TCP in high-speed networks discussed in this book will be of interest to all network engineers and researchers. ... The lists of references ... present good quality research work carried out to study the performance of TCP in computer networks, and should be used for further study." (Rinki Sharma, Computing Reviews, May, 2014)

Spis treści

Introduction.- The Transmission Control Protocol.- Modeling Incast and its Empirical Validation.- Addressing TCP Incast.- Conclusions and Future Work.

Blurb

This book addresses the need to improve TCP’s performance inside data centers by providing solutions that are both practical and backward compatible with standard TCP versions. The authors approach this challenge first by deriving an analytical model for TCP’s performance under typical data center workload traffic.  They then discuss some solutions that are designed to improve TCP performance by either proactively detecting network congestion through probabilistic retransmission or by avoiding timeout penalty through dynamic resizing of TCP segments. Experimental results show that each of techniques discussed outperforms standard TCP inside a data center.

 ·         Analyzes the reasons behind TCP performance slump inside data centers when operating under synchronized workload traffic;

·         Introduces solutions to TCP performance slump inside data centers that are both practical and backward compatible;

·         Describes mechanisms that are responsible for TCP’s reliable data transfer flow control and congestion control.