Ease Status Update

Steven Ericsson Zenith <zenith-steven@CS.YALE.EDU>
Mon, 13 Aug 90 21:48:44 GMT

          From comp.compilers

Related articles
Ease Status Update zenith-steven@CS.YALE.EDU (Steven Ericsson Zenith) (1990-08-13)
| List of all articles for this month |

Newsgroups: comp.compilers
From: Steven Ericsson Zenith <zenith-steven@CS.YALE.EDU>
Keywords: report, parallel, Ease
Organization: Compilers Central
Date: Mon, 13 Aug 90 21:48:44 GMT

We have had over 300 requests for the Ease report. I'd like to acknowledge
all the interesting and encouraging responses I've received from around the
world. Thanks everyone for the interest. I'd like to answer some of the more
general questions about Ease and its status on the net.

British and some other readers generally cannot get ftp access to Yale. The
University of Kent at Canterbury has made the report available via ftp at
Persons using guest ni-ftp should use their-username@their-hostname as
a password. They should refer to the file as: <UUCPPUBLIC>/ease
It will be kept publically available until 30/10/90. Thanks to Peter
Welch. Please notify me by email when you take a copy.

There is no working Ease compiler currently - right now I'm detailing the
language translation into ANSI standard C. Various aspects of the definition
are being tightened up - e.g. where I hand wave about recursion in the
report. *With luck* this compiler will be available freely and publicly at
the end of October (uniprocessor and shared memory multiprocessor first ..
'cos that's easier :-) ) - a report on the language translation into ANSI
standard C certainly will be available. This compiler *does not* use any
operating system facility for process creation or scheduling - so the
compiler should generate code suitable for embedded systems without
operating system support. In fact it will run code on my PC :-) Load
balancing on shared memory multiprocessors is automatic (although clearly I
don't yet know what the performance characteristics are like).

I didn't feel the Ease report was the place to get into a detailed critique
of Message Passing and Generalised Linda - I'm working on a new report, a
duo critique which gives details and examples to support the arguments made
in the Ease report. Whilst I am critical of the C-Linda combination, it
should be clear that Ease was designed with the implementation of efficient,
verifiable, high level Linda systems in mind (higher level than Ease that is).

The name? Well, Ease is a puzzle solution. The question and answer are
mutually recursive.
Q. An Abstract System in ease.
A. Ease (Abstract (A) System (S) in ease (e)s = eASe) - yea, I know, too many
New York Times crosswords :-)

Unfortunately events may distract me from the development of Ease. Grant
funding which was to have funded me (and many others) at Yale for the next
few years has unexpectedly disappeared (back there for a moment there where
thoughts of peace - so it goes...) - so I'm likely to be elsewhere come
November (CA or France I hope - maybe the UK).

I will be in CA for ten days - the whole of next week - (at HOT CHIPS) and
will be happy to visit people in the Bay Area to discuss, Ease - Linda -
Occam - Transputers - French wine - whatever.

Thanks again to all those who have shown interest and support. It's

Steven Ericsson Zenith * email: zenith@cs.yale.edu
Fax: (203) 466 2768 | voice: (203) 466 2587
Yale University Dept of Computer Science 51 Prospect St New Haven CT 06520 USA

Post a followup to this message

Return to the comp.compilers page.
Search the comp.compilers archives again.