home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.parallel
- Path: sparky!uunet!cs.utexas.edu!swrinde!gatech!hubcap!fpst
- From: kielmann@isa.informatik.th-darmstadt.de (Thilo Kielmann)
- Subject: Summary: PARMACS (long)
- Message-ID: <1992Nov5.115143.27337@news.th-darmstadt.de>
- Sender: news@news.th-darmstadt.de (The News System)
- Nntp-Posting-Host: sun24.isa.informatik.th-darmstadt.de
- Organization: TU Darmstadt
- Date: Thu, 5 Nov 1992 11:51:43 GMT
- Approved: parallel@hubcap.clemson.edu
- Lines: 121
-
- A few days ago I asked for documentation concerning the PARMACS macro
- package. Several people asked for the information I received so I decided
- to sum it up. I wish to thank the following people for their help:
-
- Mats Brorsson, Vivek Khera, Geoff Barrett, Rolf Hempel, Sanjay Raina,
- Rusty Lusk, Gail A. Alverson, Colin Wright, Peter Junglas
- (and everybody I forgot, sorry)
-
- The following lines contain the information, as I "understood" it.
- All errors are due to my ignorance.
-
-
-
- First, I cite Rolf Hempel's technical report mentioned below:
-
- ``Some years ago, scientists at Argonne National Laboratory began
- working on tools for software portability for parallel programs. In
- the beginning they were mostly interested in shared memory systems
- like the Sequent Balance. The result of that work was a package of
- macros which allowed the user to specify functions like process
- creation or synchronization in a portable way [parmacs-book].
- Moreover, these macros were implemented in a very efficient way, so
- that even with very small grained parallelism almost linear speed-up
- could be observed.''
-
- The following book contains information about this macro package:
- (Unfortunately, it seems to be out of print.)
-
- @Book{parmacs-book,
- author = "James Boyle and Ralph Butler and Terrence Disz and
- Barnett Glickfeld and Ewing Lusk and Ross Overbeek
- and James Patterson and Rick Stevens",
- title = "Portable Programs for Parallel Processors",
- publisher = "Holt, Rinehart and Winston, Inc.",
- year = 1987,
- isbn = "0-03-014153-2",
- address = "New York, NY",
- comment = "Describes the use and implementation of the PARMACS
- macros. Discuss philosophy behind message-passing,
- shared-memory, and combined MP and SM models in
- programming. Copies of the macros can be gotten
- from \mbox{\tt netlib@research.att.com} by sending
- the message {\tt send index from parmacs}."
- }
-
- This macro package may be obtained by sending email to
- netlib@ornl.gov.
- The email should consist of the following line:
-
- send index from parmacs
-
-
-
- Rusty Lusk wrote:
-
- ``Meanwhile, the Argonne group has also enhanced it (parmacs) and
- calls the new package p4. p4 is available from info.mcs.anl.gov in the
- directory pub/p4. Get the file p4t1.2.tar.Z. Documentation is included.''
-
-
- The following lines are from P4's introduction:
-
- ``p4 is a library of macros and subroutines developed at Argonne National
- Laboratory for programming a variety of parallel machines. Its predecessor
- was the m4-based "Argonne macros" system described in the Holt, Rinehart, and
- Winston book "Portable Programs for Parallel Processors, by Lusk, Overbeek, et
- al., from which p4 takes its name. The current p4 system maintains the same
- basic computational models described there (monitors for the shared-memory
- model, message-passing for the distributed-memory model, and support for
- combining the two models) while significantly increasing ease and flexibility
- of use.''
-
-
-
- Again, I cite Rolf Hempel's technical report:
-
- ``For parallel computers with local memory only there is no obvious
- way for implementing the shared memory macros. ...
- Therefore, during a visit of the author at Argonne in 1988 the
- corresponding PARMACS package for programs written in FORTRAN was
- defined and implemented on a variety of parallel systems with both
- local and shared memory.''
-
- There exists a tech report by Rolf Hempel describing these macros entitled:
-
- ``The ANL/GMD Macros (PARMACS) in FORTRAN for Portable Parallel
- Programming using the Message Passing Programming Model''
-
- This report may be requested by Rolf Hempel using his email address:
- gmap10@gmdzi.gmd.de
-
-
- Unlike the definition, the current implementation of this macro package
- is not public domain.
- Instead, it will be supported (and sold for moderate prices as Rolf
- Hempel told me) by:
- PALLAS GmbH
- Hermuehlheimer Str. 10
- D-5040 Bruehl
- Germany
-
- Tel: +49 2232 1896-21
- Fax: +49 2232 1896-29
-
- You may also contact Mr. Solchenbach at PALLAS GmbH:
- karls@pallas-gmbh.de
-
-
-
- ---
- Thilo Kielmann <kielmann@isa.informatik.th-darmstadt.de>
-
- Working at, not speaking for:
-
- University of Darmstadt
- Dept. of Computer Science
- Alexanderstr. 10
- D-6100 Darmstadt
- Germany
-
-
-