@techreport{ Fober:01c ,
title = {Real-Time IPC on a client / server model: Multiple OS performances benchmark},
author = {Dominique Fober and Yann Orlarey and Stephane Letz},
editor = {Grame},
url = {IPCBenchs.pdf},
year = {2001},
date = {2001-01-01},
booktitle = {Technical Report – 01-08-03},
abstract = {This paper presents inter processus communication (IPC) real-time performances measured on different operating systems, including GNU/Linux, Windows 98, 2000, NT 4.0 and MacOS X. The adopted point of view is based on a client / server model. The operating systems behavior and message transmission latency times are evaluated in different contexts: with one to ten clients for the server, with systems more or less busy with alternate tasks. As we wanted to measure real world performances, the benchmarks have been applied to operating systems running standard default configurations. Each time it was possible, we compared the different systems on the base of local Unix sockets communication way. But above all, we choose the most efficient communication way per system to evaluate the overall best performances that one can expect in a client / server model.},
keywords = {IPC, operating, real-time, systems},
pubstate = {published},
tppubtype = {techreport}
}
This paper presents inter processus communication (IPC) real-time performances measured on different operating systems, including GNU/Linux, Windows 98, 2000, NT 4.0 and MacOS X. The adopted point of view is based on a client / server model. The operating systems behavior and message transmission latency times are evaluated in different contexts: with one to ten clients for the server, with systems more or less busy with alternate tasks. As we wanted to measure real world performances, the benchmarks have been applied to operating systems running standard default configurations. Each time it was possible, we compared the different systems on the base of local Unix sockets communication way. But above all, we choose the most efficient communication way per system to evaluate the overall best performances that one can expect in a client / server model.
