Path: utzoo!utgpu!jarvis.csri.toronto.edu!rutgers!cs.utexas.edu!uunet!mcvax!tuvie!inst182
From: inst182@tuvie (Inst.f.Techn.Informatik)
Newsgroups: comp.realtime
Subject: Re: IEEE Tutorial (Really priority scheduled real time os)
Keywords: real-time scheduling, precedence constraints
Message-ID: <703@tuvie>
Date: 1 Jun 89 07:48:29 GMT
References: <7252@hoptoad.uucp> <34054@sgi.SGI.COM>
Reply-To: alex@honey.at (Alexander Vrchoticky)
Organization: Technical University of Vienna, EDP-Center
Lines: 29

In article <34054@sgi.SGI.COM> karsh@trifoliu.UUCP (Bruce Karsh) writes:
>In article <7252@hoptoad.uucp> gnu@hoptoad.uucp (John Gilmore) writes:
>> [stuff deleted]
>>and priority scheduling would make systems that 'seem to work' but
>>in which there is no way to engineer them so that they WILL work.
>
>  I don't agree that there is no way.
>
> [Reference to and description of rate monotonic algorithm deleted]
>
>Personally, I think this is one of the most amazing results in the field
>of real-time systems.  It is called a rate monotonic priority schedule.
>A fancier scheduler perhaps can do better, but even a conventional priority
>scheduler can do amazingly well!

Sure the result is amazing. But anyone planning to use this algorithm
really ought to look at the assumptions Liu and Layland make (page 48 
of that paper). Especially assumption 3 (Independence of tasks) does
NOT hold for any real-time system I can think of. 
The effect of relaxing these assumptions is discussed in:
	Aloysius K. Mok,
	The Design of Real-Time Programming Systems based
	on Process Models,
	Proceedings of the IEEE Real-time Systems Symposium 1984
Highly recommended ...

Alexander Vrchoticky    (Student of CS , Technical University Vienna)
alex@honey.at              
honey!alex@uunet.uu.net