From mboxrd@z Thu Jan  1 00:00:00 1970
X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me
X-Spam-Level: 
X-Spam-Status: No, score=-0.8 required=5.0 tests=BAYES_00,INVALID_DATE
	autolearn=no autolearn_force=no version=3.4.4
Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!uunet!seismo!ut-sally!husc6!cca!mirror!ishmael!inmet!stt
From: stt@inmet.UUCP
Newsgroups: comp.lang.ada
Subject: Re: RFI:Ada concurrent tasks
Message-ID: <124000006@inmet>
Date: Fri, 17-Jul-87 10:28:00 EDT
Article-I.D.: inmet.124000006
Posted: Fri Jul 17 10:28:00 1987
Date-Received: Sun, 19-Jul-87 20:35:54 EDT
References: <293@inco.UUCP>
Nf-ID: #R:inco.UUCP:-29300:inmet:124000006:000:502
Nf-From: inmet.UUCP!stt    Jul 17 10:28:00 1987
List-Id: <comp.lang.ada>



As a test case, we created and executed a program
which spawned more than 500 Ada tasks using the
Intermetrics Real-Time MVS Ada compiler on a 4341.  This compiler
uses one MVS task for each Ada task, and provides
for "true" concurrency on a multi-processor such as a 3090.
The primary performance problem involved certain
linear-list algorithms inherent in MVS dealing with
task creation.  Once the tasks were created, performance
was "fine."

Tucker Taft
c/o Intermetrics, Inc.
Cambridge, MA  02138