Path: utzoo!mnetor!tmsoft!torsqnt!news-server.csri.toronto.edu!cs.utexas.edu!sdd.hp.com!zaphod.mps.ohio-state.edu!sol.ctr.columbia.edu!emory!hubcap!ozalp
From: ozalp@dm.unibo.it (Ozalp Babaoglu)
Newsgroups: comp.parallel
Subject: Re: simple coarse-grained parallel programs
Message-ID: <12337@hubcap.clemson.edu>
Date: 18 Dec 90 19:56:25 GMT
Sender: fpst@hubcap.clemson.edu
Reply-To: ozalp@dm.unibo.it (Ozalp Babaoglu)
Organization: Department of Mathematics, University of Bologna, Italy
Lines: 77
Approved: parallel@hubcap.clemson.edu

In article <12179@hubcap.clemson.edu> anand@top.cis.syr.edu (Rangachari Anand) writes:
>Using a group of workstations connected on a LAN  as a prallel
>computer is an idea that is continually being rediscovered. Given the 
>high speeds of the available workstations, this is clearly an Idea Whose 
>Time Has Come.

add one more to the list:  Paralex, being developed here at the university
of Bologna, Italy.  i am enclosing a brief description of the scope and
goals of paralex.  for further information, please contact me by e-mail.

--

		Paralex:  An Environment for Parallel Programming
			in Distributed Systems


One of the many advantages of distributed systems is their ability to
execute several computations on behalf of a single application in
parallel, thus improving performance.  In fact, at a certain level of
abstraction, there is little difference between a distributed system
and a losely-coupled multiprocessor computer.  We cannot, however, treat
distributed systems as if they were uniform multiprocessor parallel
machines due to the following characteristics:

	o  High latency, low bandwidth communication

	o  Presence of heterogeneous processor architectures

	o  Communication link and processor failures

	o  Multiple independent administrative domains.

Thus, if we can address these issues, a distributed computing resource
such as a collection of workstations could be viewed and used as if it
were a poor man's ``super computer.''  To make a distributed system
suitable for long-running parallel computations, support must be
provided for fault tolerance.  Many hours of computation can be wasted
not only if there are hardware failures, but also if one of the
processors is turned off, rebooted or disconnected from the network.
Given that the components of the system (workstations) may be under
the control of several independent administrative domains (typically a
single individual who ``owns'' the workstation), these events are much
more plausible and frequent than real hardware failures.

It is our thesis that reasonable technologies already exist to
address the problems related to distribution, communication and fault
tolerance of applications in distributed systems.  What remains a
challenge is the task of {\em programming} reliable applications that
can benefit from the parallelism and fault tolerance that distributed
systems have to offer.  The Paralex Project has been undertaken to
explore the extent to which the programmer can be liberated from the
complexities of distributed systems.  Our goal is to realize an
environment that will encompass all phases of the programming activity
and provide automatic support for distribution, fault tolerance and
heterogeneity in distributed and parallel applications.

Paralex makes extensive use of graphics for expressing computations,
monitoring, controlling execution and debugging.  In fact, the programming
paradigm supported by Paralex is best suited for parallel computations
that can be viewed as collages of ordinary sequential programs.  The
interdependencies and data flow relations between computations in a
parallel program are expressed in a natural way using a graphical
notation.  In the limit, interesting new parallel programs can be
``programmed'' by reusing existing sequential software and without
having to write a single line of traditional code.

A prototype of Paralex is being built at the Department of Mathematics,
University of Bologna, Italy for a network of Sun-3, Sun-4 and Vax
architecture workstations running Unix.  The graphical interface is
based on X-Windows and the Open Look Intrinsics.  Automatic support
for fault tolerance is achieved through the ISIS toolkit.

--
Ozalp Babaoglu				       E-mail:	ozalp@dm.unibo.it
University of Bologna, Dept. of Mathematics
Piazza di Porta S. Donato, 5		       TEL:	+39 51 354430
40127 Bologna  (ITALY)			       FAX:	+39 51 354490