Subject: SDL-News: Performance and Time
From: Andreas Mitschele_Thiel (mitsch#immd7.informatik.uni-erlangen.de)
Date: Thu Nov 13 1997 - 14:07:48 GMT
The originator of this message is responsible for its content.
-----From Andreas Mitschele_Thiel <mitsch#immd7.informatik.uni-erlangen.de> to sdlnews -----
Dear SDL and MSC folks,
In order to foster the integration of performance and time aspects into SDL,
I put up some preliminary guidelines (see below) draw from the discussion
at the Lutterworth meeting in October.
In order to further elaborate on the issue, it is proposed that the next
Q.6/10 expert meeting will be held in conjunction with a workshop on
performance and time in SDL/MSC.
PRELIMINARY GUIDELINES FOR THE INTEGRATION OF PERFORMANCE AND TIME ASPECTS
IN SDL AND MSC
(Results of the Joint Q6,9/10 Meeting in Lutterworth, Oct. 15, 1997)
Currently, SDL and MSC do not support various aspects related to performance
and time (see TDL620 for details).
In order to deal with performance and time aspects in SDL and MSC
additional information has to be associated with the SDL and/or MSC
specification. The most important information are:
- the arrival process, describing the signals arriving at the system,
- the resource requirements of the parts of the specification,
- the available resources,
- the mapping of the parts of the specification on the resources, and
- the sensors to specify which performance information is probed.
The meeting agreed that performance and time are very important aspects
and should be dealt with in the context of SDL and MSC.
In addition, some (weak) agreement could be achieved concerning the following:
All information typically kept in the implementation description, i.e.
the available resources and the mapping on the resources, should remain
outside of SDL and MSC.
However, a standardized approach to specify this information is needed in
order to support the application of different performance tools.
No agreement could be reached concerning formal specification of the remaining
information. Tool vendors tend to prefer SDL, the MSC community would like
associate the information with MSCs. A third alternative is to keep the
information completely separate from SDL and MSC and to rely on the tools
to edit and display the respective information within the SDL and/or MSC
In the case of language extensions to SDL and/or MSC, it is not clear
whether the extensions should be annotational (confined in comments of the
original language) or real language extensions.
However, there does not seem to be a strong objection to real language
Up to date it is also open where to place the performance and time extensions
in the standard documents. Alternatives are the Z.100 and/or Z.120 standard
itself, an appendix to the standards or a new separate standard adjunct with
the current standards (i.e. as it is the case with ASN.1 and CIF).
-- Dr. Andreas Mitschele-Thiel University of Erlangen-Nuremberg Phone: +49-9131-85-7932 or -7411 IMMD VII Fax: +49-9131-85-7409 Martensstr. 3 Email: mitsch#informatik.uni-erlangen.de 91058 Erlangen Germany
-----End text from Andreas Mitschele_Thiel <mitsch#immd7.informatik.uni-erlangen.de> to sdlnews ----- For help, email "majordomo#sdl-forum.org" with the body of your email as: help or (iff this does not answer your question) email: owner-sdlnews#sdl-forum.org
This archive was generated by hypermail 2a23 : Sun Jun 16 2013 - 10:41:40 GMT