DocumentCode :
1433561
Title :
Application-level performance prediction tools for network-based svstems
Author :
Bulterman, Dick C. A. ; Manolis, E.
Author_Institution :
Dept. of Eng., Brown Univ., Providence, RI, USA
Volume :
1
Issue :
3
fYear :
1987
fDate :
7/1/1987 12:00:00 AM
Firstpage :
6
Lastpage :
12
Abstract :
A description is given of a network simulation system that is used to monitor process-to-process efficiency in a special purpose network developed for real-time applications. The tool, which is called VISA (Visually Integrated System Analyzer), uses information about application-specific interactions to predict how well a network-based implementation of that application will perform. VISA is part of a larger collection of system-level engineering tools that are being developed under the CASE (Computer-Aided System Engineering) environment. The VISA approach partitions a graph-based application system into a set of interconnected network clusters, each cluster able to support up to 200 processing nodes, and can be used to determine if a particular network partitioning can support the requirements of the application graph. The nature of the applications that the VISA system supports is briefly discussed, as well as the system itself, along with a short example of its use, and how it is logically structured. The structure of the CASE environment is discussed, and it is shown how the VISA system is integrated into it.
Keywords :
computer networks; computerised signal processing; digital simulation; performance evaluation; CASE; VISA; Visually Integrated System Analyzer; application level performance prediction tools; application-specific interactions; graph-based application system; interconnected network clusters; network simulation system; network-based systems; process-to-process efficiency; real-time applications; special purpose network; system-level engineering tools; Computer aided software engineering; Computer architecture; Digital signal processing; Hardware; Monitoring; Program processors; Time measurement;
fLanguage :
English
Journal_Title :
Network, IEEE
Publisher :
ieee
ISSN :
0890-8044
Type :
jour
DOI :
10.1109/MNET.1987.6434205
Filename :
6434205
Link To Document :
بازگشت