From owner-hpff Fri Mar 1 04:03:35 1996 Received: (from daemon@localhost) by cs.rice.edu (8.7.1/8.7.1) id EAA14950 for hpff-out; Fri, 1 Mar 1996 04:03:35 -0600 (CST) Received: from rising.irisa.fr (rising.irisa.fr [131.254.41.16]) by cs.rice.edu (8.7.1/8.7.1) with SMTP id EAA14944 for ; Fri, 1 Mar 1996 04:03:26 -0600 (CST) Received: from yapuka.irisa.fr (yapuka.irisa.fr [131.254.42.19]) by rising.irisa.fr (8.6.12/8.6.9) with ESMTP id LAA27861; Fri, 1 Mar 1996 11:02:37 +0100 Message-Id: <199603011002.LAA27861@rising.irisa.fr> To: hpff@cs.rice.edu, ptools@monroe.llnl.gov, prs@uvsq.fr Subject: hpff: High Performance Fortran Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 01 Mar 1996 11:02:36 +0100 From: Jean Louis Pazat Sender: owner-hpff Precedence: bulk --------------------------------------------------------------------------- hpff@cs.rice.edu is a mailing list for announcements related to High Performance Fortran. Instructions for adding or deleting yourself from this list appear at the bottom of this message. --------------------------------------------------------------------------- Dear (HPF) tool developers, A list of HPF compilers can be accessed at http://www.irisa.fr/pampa/HPF/compilers.html If you are working on an HPF compiler and if you are not in this list send a mail with the address of your Web page to pazat@irisa.fr. You will be added to this list. A survey about HPF related tools is under construction at http://www.irisa.fr/pampa/HPF/survey.html If you are willing to contribute to this survey send a mail to pazat@irisa.fr with a brief description of your tool Information about HPF in Europe is available from: http://www.irisa.fr/pampa/HPF/HPF.html Thanks for your cooperation. Dr. Jean-louis PAZAT. PAMPA Team IRISA Phone : +33-99-84-72-14 | Fax: +33-99-84-71-71 e-mail : pazat@irisa.fr | Web: http://www.irisa.fr/prive/pazat/ address: IRISA Campus Universitaire de Beaulieu 35042 RENNES CEDEX FRANCE --------------------------------------------------------------------------- To (un)subscribe to this list, send mail to hpff-request@cs.rice.edu. Leave the subject line blank, and in the body put the line (un)subscribe --------------------------------------------------------------------------- From owner-hpff Thu Mar 7 14:08:53 1996 Received: (from daemon@localhost) by cs.rice.edu (8.7.1/8.7.1) id OAA28778 for hpff-out; Thu, 7 Mar 1996 14:08:53 -0600 (CST) Received: from pacific.pgroup.com (pacific.pgroup.com [192.124.124.8]) by cs.rice.edu (8.7.1/8.7.1) with SMTP id OAA28761 for ; Thu, 7 Mar 1996 14:08:37 -0600 (CST) Received: (from miles@localhost) by pacific.pgroup.com (8.6.12/8.6.11) id MAA24720 for hpff@cs.rice.edu; Thu, 7 Mar 1996 12:08:24 -0800 Date: Thu, 7 Mar 1996 12:08:24 -0800 From: Doug Miles Message-Id: <199603072008.MAA24720@pacific.pgroup.com> To: hpff@cs.rice.edu Subject: hpff: Cray / PGI / HPF X-Sun-Charset: US-ASCII Sender: owner-hpff Precedence: bulk --------------------------------------------------------------------------- hpff@cs.rice.edu is a mailing list for announcements related to High Performance Fortran. Instructions for adding or deleting yourself from this list appear at the bottom of this message. --------------------------------------------------------------------------- Cray/Media: Steve Conway, 612/683-7133 The Portland Group: Denney Cole, 503/682-2806 The Bernhardt Agency: Cheri Maniscalco, 503/226-6452 CRAY RESEARCH AND THE PORTLAND GROUP FINALIZE AGREEMENTS FOR HPF ON CRAY SYSTEMS, PGI TO DEVELOP NEW MERGED PROGRAMMING PRODUCT BASED ON CRAY/PGI TECHNOLOGY Eagan, Minn., March 7, 1996 -- Cray Research, Inc. (NYSE:CYR) and The Portland Group, Inc. (PGI) announced today that they have finalized a development and reseller agreement to initially make PGI's pghpf(tm) High Performance Fortran (HPF) compiler available on Cray(R) systems and for PGI to develop and support a merged programming product based on Cray's CRAFT flexible programming model and PGI's pghpf, creating the most complete and powerful implicit programming model available on the market, according to the two companies. Late last year the two companies announced their intent to make PGI's HPF product available on Cray systems. Today's announcement extends that intended relationship with a reseller arrangement and an important development program to combine Cray's programming model with PGI's technology to create a comprehensive programming solution for both Cray and PGI customers. The combined product is expected to be available in late 1996. The agreement is important because implicit programming models are easiest to use on today's highly scalable, distributed memory systems, and the combined model announced today is the most capable in the world, the companies said. Under terms of the development agreement, to ensure cross- platform portability, the merged programming model will be supported by PGI on a variety of parallel systems, including all Cray Research shared memory systems, as well as the scalable CRAY T3D and CRAY T3E distributed memory supercomputers. Under the terms of the reseller agreement, Cray will resell PGI's HPF, as well as the merged programming product on all Cray(R) systems. According to Mike Booth, vice president of Cray's software division, "We've been working with PGI for several months toward a concrete plan and schedule for integrating the HPF and CRAFT programming models. We are pleased to announce that we'll be providing Cray users with an HPF environment that is consistent across all of our product lines, provides maximum power and flexibility on both shared- and distributed-memory systems, and is based on PGI's industry- leading HPF product." According to Douglas Miles, director of marketing at PGI, "We're very pleased Cray will be offering our HPF products directly to their customers. Expanding the relationship to incorporate the power of Cray's CRAFT model will be of substantial benefit to many existing Cray customers, as well as PGI's customers. CRAY T3D/T3E users will be able to migrate their applications toward HPF compatibility with the freedom of using the CRAFT model where appropriate and the assurance that their code will continue to be portable across a variety of parallel systems." Under the separate development agreement, PGI will make enhancements to its HPF product to incorporate the functionality of Cray's CRAFT programming model. Cray Research developed the CRAFT parallel programming model to provide CRAY T3D(tm) customers with a flexible model that combines the functionality of several programming styles -- data parallel, work sharing, and message-passing. The key components of the CRAFT model will be embedded within a new HPF standard-conforming extrinsic type called EXTRINSIC(HPF_CRAFT) to provide a multi-threaded execution environment. This combined Cray/PGI solution will give programmers all the flexibility of CRAFT incorporated in a standard fashion within a portable language. HPF is the leading implicit parallel programming model for shared- and distributed-memory parallel systems. It is the first widely accepted programming language suitable for writing portable applications in the data-parallel programming style. PGI provides the leading commercial implementation of HPF on a variety of parallel systems. Cray Research provides the leading high-performance tools and services to help solve customers' most challenging problems. PGI is a leading independent vendor of software compilers and tools for parallel computing. PGI provides high performance, retargetable, production quality compilers and software development tools to the high performance and parallel computing industries. # # # --------------------------------------------------------------------------- To (un)subscribe to this list, send mail to hpff-request@cs.rice.edu. Leave the subject line blank, and in the body put the line (un)subscribe --------------------------------------------------------------------------- From owner-hpff Thu Mar 7 16:39:43 1996 Received: (from daemon@localhost) by cs.rice.edu (8.7.1/8.7.1) id QAA06996 for hpff-out; Thu, 7 Mar 1996 16:39:43 -0600 (CST) Received: from theory.tc.cornell.edu (THEORY.TC.CORNELL.EDU [132.236.98.174]) by cs.rice.edu (8.7.1/8.7.1) with SMTP id QAA06991 for ; Thu, 7 Mar 1996 16:39:36 -0600 (CST) Received: (from presberg@localhost) by theory.tc.cornell.edu (8.6.9/8.6.6) id RAA01836; Thu, 7 Mar 1996 17:39:20 -0500 Date: Thu, 7 Mar 1996 17:39:20 -0500 Message-Id: <199603072239.RAA01836@theory.tc.cornell.edu> From: David Presberg To: Doug Miles Cc: hpff@cs.rice.edu Subject: re: hpff: Cray / PGI / HPF In-Reply-To: <199603072008.MAA24720@pacific.pgroup.com> References: <199603072008.MAA24720@pacific.pgroup.com> cc: presberg@tc.cornell.edu Sender: owner-hpff Precedence: bulk --------------------------------------------------------------------------- hpff@cs.rice.edu is a mailing list for announcements related to High Performance Fortran. Instructions for adding or deleting yourself from this list appear at the bottom of this message. --------------------------------------------------------------------------- Doug -- I think it would have been better to have just pointed all of us at some online copy of the Press Release (e.g., URL, anonymous FTP). -- Pres - ------------------------------------------------------------------------ - David L. Presberg, Parallel Computing Specialist, CTC/ST/Tools - Cornell Theory Center, 740 Frank H.T. Rhodes Hall, Cornell University, - Ithaca, NY 14853-3801, USA 607-254-8861 presberg@tc.cornell.edu - Nickname: Pres - ------------------------------------------------------------------------ --------------------------------------------------------------------------- To (un)subscribe to this list, send mail to hpff-request@cs.rice.edu. Leave the subject line blank, and in the body put the line (un)subscribe --------------------------------------------------------------------------- From owner-hpff Sun Mar 10 18:04:37 1996 Received: (from daemon@localhost) by cs.rice.edu (8.7.1/8.7.1) id SAA06187 for hpff-out; Sun, 10 Mar 1996 18:04:37 -0600 (CST) Received: from ion3.ionet.net (ion3.ionet.net [204.96.200.8]) by cs.rice.edu (8.7.1/8.7.1) with SMTP id SAA06182 for ; Sun, 10 Mar 1996 18:04:30 -0600 (CST) Received: from erehwon (tsip145.ionet.net [206.28.164.154]) by ion3.ionet.net (8.6.12/8.6.12) with SMTP id RAA17309; Sun, 10 Mar 1996 17:57:38 -0600 X-Mailer: InterCon TCP/Connect II 2.2.1 MIME-Version: 1.0 Message-Id: <9603101636.AA32936@erehwon> Date: Sun, 10 Mar 1996 16:36:32 -0500 From: "Jerrold L. Wagener" To: emili@nepal.udg.es Cc: sc22wg5@dkuug.dk, hpff@cs.rice.edu Subject: hpff: Re: (x3j3.1996-73) NSS Content-Type: Multipart/Mixed;boundary=part_AD68B58F0006123100000003 Sender: owner-hpff Precedence: bulk --------------------------------------------------------------------------- hpff@cs.rice.edu is a mailing list for announcements related to High Performance Fortran. Instructions for adding or deleting yourself from this list appear at the bottom of this message. --------------------------------------------------------------------------- --part_AD68B58F0006123100000003 Content-Transfer-Encoding: Quoted-Printable Content-Type: Text/Plain; charset=ISO-8859-1 Content-Disposition: Inline Dr. Besalu, thank you for your note about NSS. By copy of this reply I will forward = it to both the international Fortran standards group (SC22/WG5) and = the group instrumental in developing FORALL parallelism (HPFF). I'm afraid you're too late for Fortran 95, as that has essentially gone = to press, but WG5 is calling now for proposed requirements for Fortran = 2000. You might want to contact Dr. Miles Ellis (miles.ellis@= etrc.ox.ac.uk) with regard to submiting such a proposal. The procedure = is that such proposals are submited by national bodies to WG5 - Dr. = Ellis can put you in touch with the appropriate contact in Spain. I'm sure that those individuals in X3J3, WG5, and HPFF interested in = high performance and in better mouse traps, so to speak, will look at = NSS on the web page you provided. But there is no guarantee that any = will champion its cause; as with most anything, the best approach is = to champion it yourself within these various groups. Thanks again for your note, and let me know if there is anything more I = can do to assist you in this matter. Jerry Wagener chair, X3J3 jwagener@ionet.net jwagener@cs.uoknor.edu =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D > =20 > I know you are related to the standards of Fortran 95. I'm a=20 > programmer using this language and I have developed an algorithm=20 > called Nested Summation Symbol (NSS) which is close related to the=20 > FORALL sentence. Although the NSS is more general.=20 > It is also parallellizable. I think it is undoubtedly a better choice=20 > than the FORALL sentence. A NSS symbolizes AN ARBITRARY number=20 > of summation symbols or DO LOOPS. This is the main difference between=20 > the FORALL sentence and the NSS. The FORALL sentence allows only=20 > the representation of a FIXED number of summation symbols or loops.=20 > =20 > I wonder if you may be interested into this new algorithm. In this=20 > case you can contact me via E-mail. Also I invite you to see my WWW=20 > page at the location http://stark.udg.es/cat/emili.html.=20 > There you can find more information about the NSS: code,=20 > examples, references, and more.=20 > =20 > In some papers I have suggested the implementation of the NSS into=20 > the high level languages, such as Fortran or C.=20 > =20 > Please, do not hesitate to visit my WWW page.=20 > =20 > Perhaps the NSS construct may be worth to be included into the=20 > Fortran 95 standard.=20 > =20 > Please, give me an answer! (at least in order to know if this message=20 > has been read) Thanks for your attention.=20 > =20 > Bye ! =20 > =20 > E. Besal=A3=20 > =20 --part_AD68B58F0006123100000003 Content-Type: Text/Plain; charset=US-ASCII Content-Disposition: Inline ========================================== Jerrold L. Wagener jwagener@ionet.net 6 E 5th, Suite 308 918-592-3023 Tulsa, OK 74103 --part_AD68B58F0006123100000003-- --------------------------------------------------------------------------- To (un)subscribe to this list, send mail to hpff-request@cs.rice.edu. Leave the subject line blank, and in the body put the line (un)subscribe --------------------------------------------------------------------------- From owner-hpff Mon Mar 11 06:20:31 1996 Received: (from daemon@localhost) by cs.rice.edu (8.7.1/8.7.1) id GAA18851 for hpff-out; Mon, 11 Mar 1996 06:20:31 -0600 (CST) Received: from sescva (sescva.esc.edu [138.116.1.1]) by cs.rice.edu (8.7.1/8.7.1) with SMTP id GAA18844 for ; Mon, 11 Mar 1996 06:20:27 -0600 (CST) Date: Mon, 11 Mar 1996 07:23:14 -0500 Message-Id: <96031107231418@sescva.esc.edu> From: dswift@sescva.esc.edu (J. DAVID SWIFT) To: hpff@cs.rice.edu Subject: hpff: Message for Ricky Swift X-VMS-To: SMTP%"hpff@cs.rice.edu" X-VMS-Cc: DSWIFT Sender: owner-hpff Precedence: bulk --------------------------------------------------------------------------- hpff@cs.rice.edu is a mailing list for announcements related to High Performance Fortran. Instructions for adding or deleting yourself from this list appear at the bottom of this message. --------------------------------------------------------------------------- HPFF Correspondents, I am trying to locate Ricky Swift's email address. If Ricky himself or anyone else knows Rick's address/phone please send it to me. Thanks. ---------------------------------------------------------------- J. David Swift 10084 Hooker Hill Road INTERNET:DSWIFT@sescva.esc.edu Perrysburg, NY 14129-9715 (716) 532-6022 [voice & fax] Hope to hear from you soon. ------------------------------------------------------------------ --------------------------------------------------------------------------- To (un)subscribe to this list, send mail to hpff-request@cs.rice.edu. Leave the subject line blank, and in the body put the line (un)subscribe ---------------------------------------------------------------------------