Potential Performance Bottleneck in Linux TCP

Document #:

CS-doc-1648-v2

Document type:

Technical Note

Submitted by:

Matt Crawford

Updated by:

Matt Crawford

Document Created:

26 Jul 2006, 11:49

Contents Revised:

11 May 2007, 13:36

Metadata Revised:

11 May 2007, 13:45

Abstract:

TCP is the most widely used transport protocol on the Internet today. Over the years, espe-cially recently, due to requirements of high bandwidth transmission, various approaches have been proposed to improve TCP performance. The Linux 2.6 kernel is now preempti-ble. It can be interrupted mid-task, making the system more responsive and interactive. However, we have noticed that Linux kernel preemption can interact badly with the per-formance of the networking subsystem. In this paper we investigate the performance bottleneck in Linux TCP. We systematically describe the trip of a TCP packet from its in-gress into a Linux network end system to its final delivery to the application; we study the performance bottleneck in Linux TCP through mathematical modeling and practical ex-periments; finally we propose and test one possible solution to resolve this performance bottleneck in Linux TCP.

Files in Document:

• Final version of paper for publication (Potential Performance Bottleneck in Linux TCP.PDF, 3.8 MB)

Topics:

• Shared Systems and Services:Unix/Linux Servers:Apache
• High Performance Computing

Authors:

• Matt Crawford
• Wenji Wu

Keywords:

Linux TCP network performance scheduling

Publication Information:

International Journal of Communication Systems, John Wiley & Sons
http://dx.doi.org/10.1002/dac.872