Re: Beowulf + E ? Marc Stiegler (marcs@skyhunter.com)
Fri, 7 May 1999 08:56:39 -0700

The other way E looks to work very nicely is as the "stitching material" for programs written in Java, which will probably be made to run fast before E is made to run fast. I have been looking at my Reversi program (posted on the Web for a couple years now) with an eye to a distributed-computation version--anyone on the Web could start up a copy of the Reversi Computation system and have their computer become a part of the analysis, with an eye to doing a complete solution of the game (alpha-beta-minimax to 60 ply, at which point the game is over). It looks alluringly straightforward to wrap the BoardEvaluator Java class with enough E machinery to just manage the cross-computer communication. Of course, going all the way for the performance freaks in these days before Java runs fast, you'd do BoardEvaluator in C (or assembler! :-) The cost of even current E would be negligible.

--marcs

-----Original Message-----
From: Mark S. Miller <markm@caplet.com>
To: Chip Morningstar <chip@communities.com> Cc: e-lang@eros.cis.upenn.edu <e-lang@eros.cis.upenn.edu> Date: Thursday, May 06, 1999 11:55 PM
Subject: Re: Beowulf + E ?

>At 02:49 PM 5/6/99 , Chip Morningstar wrote:
>>Bill Frantz and I were chatting over lunch yesterday and it occurred to me
to
>>wonder if you have investigated/looked into/heard of the community of
folks
>>doing parallel computation with what they call Beowulf clustering (see,
for
>>example, www.beowulf.org).
>
>Nothing deep. Just a few casual conversations.
>
>
>>...
>>Second, I'm thinking about the model presented in _Crossing The Chasm_.
>Perhaps
>>we should think about more consciously positioning E as the Perl or Visual
>>Basic of distributed systems: its job is not to run fast, its job is to
enable
>>unsophisticated programmers of average competence to quickly and fairly
easily
>>whip together boring but valuable small-to-medium sized (distributed)
>>applications. Being the hardcore hacker early adopter types that we are,
we
>>*like* to think of what we are doing here as edge-of-the-art stuff, but it
may
>>be more advantageous to posture as the second wave of technology, making
this
>>class of systems accessible to the mainstream.
>
>I quite agree with this last paragraph. The difference between Joule & E
is
>primarily E's effort to be immediately understandable and familiar to a
>large number of Perl, Python, & VB level programmers; and to enable them to
>casually create working distributed applications, without needing a deep
>knowledge of distributed system foo.
>
>However, the Internet as a whole is a much bigger laboratory than Beowulf
>clusters, and there's a lot more payoff for a lot more people if they can
>target their distributed app development at the Open Internet. With E,
this
>shouldn't be harder than targeting a closed Beowulf cluster.
>
>A good example is MarcS' upcoming Secureit-EChat -- a truly secure 2-person
>distributed chat program in 5 pages of E code, about 3 of which are user
>interface. Watch http://www.skyhunter.com/marc.html
>
>Of course, if a Beowulf-based opportunity to advance the project presented
>itself, great!
>
> Cheers,
> --MarkM
>