✦ LIBER ✦

PVM communication performance over an ATM MAN

✍ Scribed by Giulio Iannello; Stefano Russo

Publisher: Elsevier Science
Year: 1997
Tongue: English
Weight: 699 KB
Volume: 43
Category: Article
ISSN: 1383-7621
DOI: 10.1016/s1383-7621(96)00080-x

No coin nor oath required. For personal study only.

✦ Synopsis

This paper presents results of the performance tests of the PVM system over an ArM Metropolitan Area Network (MAN). PVM (Parallel Virtual Machine) is a communication package, which is considered currently a de facto standard for the development of high-performance distributed applications. ATM (Asynchronous Transfer Mode) is the emerging cell-switching technology for high-speed networks. This work is part of the experiments of heterogeneous network computing in the framework of the Neanet Project. Neanet is one of the first public high-speed networks based on ATM. The ultimate goal of the e.iperlments is to assess the effectiveness of ATM geographical networks for high-performance computing. Besides allowing comparison with local area networks, the YVM performance figures are useful per se in the design and in the predictive performance evaluation of distributed applications on ATM networks.

📜 SIMILAR VOLUMES

Performance analysis of cell discarding

Performance analysis of cell discarding techniques for best effort video communications over ATM networks

✍ Ahmed Mehaoua; Raouf Boutaba 📂 Article 📅 1998 🏛 Elsevier Science 🌐 English ⚖ 542 KB

With increasing interest in the transmission of audio-visual applications over ATM best effort services, efficient video-oriented control mechanisms for improving the video quality in the presence of loss have to be designed. In this paper, we propose and evaluate two new slice-based discard schemes

An empirical study of TCP/IP performance

An empirical study of TCP/IP performance over FR/ATM service interworking

✍ Isil Sebüktekin 📂 Article 📅 1998 🏛 John Wiley and Sons 🌐 English ⚖ 209 KB

A key result of these empirical studies is the requirement to configure adequate buffer space in customer premises equipment, which has a DS1 bottleneck itself or which sends data to a DS1 bottleneck in the FR/ATM interworking equipment. As long as such care is exercised, the DS3-to-DS1 rate mismatc