DOCSLIB.ORG

Netlib and NA-Net: Building a Scientific Computing Community Jack Dongarra Gene H

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Netlib and NA-Net: Building a Scientific Computing Community Jack Dongarra Gene H Netlib and NA-Net: Building A Scientific Computing Community Jack Dongarra Gene H. Golub Eric Grosse Cleve Moler More IEEE Annals of the History of Computing, Volume 30, Number 2, April-June 2008, pp. 30-41 (Article) Published by IEEE Computer Society For additional information about this article http://muse.jhu.edu/journals/ahc/summary/v030/30.2.dongarra.html Access Provided by University of Tennessee @ Knoxville at 06/21/11 2:06PM GMT Netlib and NA-Net: Building A Scientific Computing Community Jack Dongarra University of Tennessee, ORNL, and University of Manchester Gene H. Golub (Deceased 16 November 2007) Eric Grosse Google Cleve Moler MathWorks Keith Moore University of Tennessee Two resources evolved in the early 1980s to serve the needs of the scientific computing community. These resources were Netlib, a software repository that facilitated distribution of public-domain software, and NA-Net, a file of analysts’ contact information that eventually supported an online directory and newsletter digest. The authors who created Netlib and NA-Net explain the history of these resources and their continuing impact. With today’s effortless access to open developed resources filling different func- source software and data via a high-speed tions but which were both aimed at serving Internet and responsive search engines, it is the scientific computing community. These hard to remember just how different the life resources, still active today, were Netlib and of computational scientists was in the 1970s NA-Net. and early 1980s. At the time, computing Netlib repositories—collections of mathe- largely took place in a mainframe world, matical software, papers, and databases—were typically supported by one of the few initially established in 1984 by Eric Grosse at commercial numerical libraries installed by Bell Labs and Jack Dongarra at Argonne computer center staff. Since this situation National Laboratory. Based on a suggestion was imperfect (the existing software did not from Gene Golub to build a repository of solve all of scientific computing’s needs), research software, Dongarra and Grosse de- scientists were forced to write or borrow signed the first Netlib repository. Each repos- additional software. There were a few exem- itory made a collection of high-quality math- plary libraries such as EISPACK in circulation, ematical software available via electronic and a larger body of Fortran programs of mail—the Bell Labs server provided access via variable quality. Obtaining them was a uucp (Unix to Unix CoPy) protocols, while the bothersome process, however, that involved Argonne server made the collection accessible leveraging personal contacts, government via IP (Internet Protocol). The Netlib collection bureaucracies, negotiated legal agreements, quickly spread to mirror servers around the and the expensive and unreliable shipping of world. As the Internet became ubiquitous, 9-track magnetic tapes or punch card decks. other access methods were added. Although There had to be a better way. That better way the electronic mail interface is still supported, manifested itself as two different, separately most access today is via the World Wide Web. 30 IEEE Annals of the History of Computing Published by the IEEE Computer Society 1058-6180/08/$25.00 G 2008 IEEE Prior to Netlib (short for ‘‘network library’’), message consisting of the text ‘‘send dqag software codes had been shared by interest from quadpack’’ sent to either research!netlib groups such as SHARE since the mid-1950s, but (uucp)ornetlib@anl-mcs (Internet) would such groups tended to be limited by member- result in a reply consisting of the ‘‘dqag’’ ship or personal relationships. By widening subroutine from the ‘‘quadpack’’ package, the distribution of public-domain numerical along with any additional routines needed to software, Netlib gave greater leverage to early use that subroutine. open source efforts and helped establish the At the time Netlib was introduced, public- trend of software distribution via the network. domain software was chiefly distributed by Netlib remains popular today. physical media such as magnetic tapes being In a separate development that was also sent through the postal service. This was aimed at the scientific computing communi- difficult not only because of the time and ty, by the time Netlib was established, an ‘‘na expense involved with handling physical list’’ (na for ‘‘numerical analysis’’) of email media, but also because of the lack of any addresses had already been maintained for widely used standards for writing magnetic several years by Gene Golub, then chair of the tapes. Each computing platform wrote tapes Computer Science Department at Stanford in its native format; and because of differ- University. By 1983, this list was being used to ences in word size, record size, and organi- provide an electronic mail forwarding service zation of the data on tape, reading foreign to numerical analysis specialists. Mail to na. tapes could be quite difficult and labor lastname@su-score would be forwarded to the intensive. Fortran programs were formatted list member with that last name. An email with sequence numbers in columns 73–80 broadcast facility was also provided: mail sent and blank fill, which was painfully ineffi- to na@su-score would be forwarded to every- cient in days of dial-up lines before compres- one on the list. By February 1987, this sion was prevalent. broadcast facility had evolved into a moder- Of course, the Arpanet had been in exis- ated email digest, which soon became a tence for several years by that time, and weekly electronic newsletter. A ‘‘white pages’’ software was commonly shared over the database and a World Wide Web interface Arpanet’s File Transfer Protocol (FTP), but the were eventually added, resulting in the set of Arpanet served a limited community, and the services provided by NA-Net today. The NA- Internet protocols adopted by the limited- Net remains a widely used and valuable commercial-use Arpanet in 1981 were only resource for the numerical analysis commu- beginning to see wider use. Software could also nity. The NA-Digest is one of the oldest be uploaded to or downloaded from ‘‘bulletin electronic periodicals, and its popularity board systems’’ or BBSs, but this could entail continues to grow. significant long-distance telephone charges to In this article, which describes the early peruse distant servers. A picture of the techni- history of these two resources and the ratio- cal environment of the early Internet, uucp, nale behind the system architectures chosen, Bitnet, and separate systems like AOL with we assess what parts worked well and which gateway size limits due to dial-up and restart- did not. from-beginning error recovery has been docu- mented elsewhere.1,2 Netlib By contrast, Arpanet electronic mail sys- The concept behind Netlib was to facili- tems standardized very early on a common tate quick, easy, and efficient on-demand message format to be used across all platforms, access to useful public-domain computation- and uucp mail used a similar format. Bitnet had al software of interest to the scientific some different conventions, but adequate computing community. The mechanism gateways existed. Costs of email were absorbed chosen for distribution of this software was into budget overheads, so experimental use electronic mail. Initially, there were two could flourish without formalities. Collective- repositories: one at Bell Labs and the other ly, these conditions made electronic mail a at Argonne National Laboratory. The Bell superior medium for the exchange of source Labs server provided access via uucp proto- code, at least for small files, and Netlib made cols, while the Argonne server made the effective use of it. Unknown to us at the time, collection accessible via Internet mail. To silicon chip designs were being sent off for request files from either server, a user would fabrication by email in a service known as send one or more commands as message text MOSIS (Metal Oxide Semiconductor Imple- to that server’s email address. For instance, a mentation Service).3 Another early informa- April–June 2008 31 Netlib and NA-Net: Building A Scientific Computing Community Definition of Terms AMS—The American Mathematical Society (AMS) is creation of user interfaces, and interfacing with pro- an association of professional mathematicians dedicated grams in other languages. to the interests of mathematical research and scholar- MOSIS—MOSIS (Metal Oxide Semiconductor Im- ship through various publications and conferences. plementation Service) is probably the oldest (1981) BLAS—Basic Linear Algebra Subprograms (BLAS) are integrated circuit (IC) foundry service and one of the standardized application programming interfaces for first Internet services other than supercomputing subroutines to perform basic linear algebra operations services and basic infrastructure such as email or FTP. such as vector and matrix multiplication. MINPACK—A library of Fortran subroutines for the EISPACK—A software library for numerical computa- solving of systems of nonlinear equations, or the least- tion of eigenvalues and eigenvectors of matrices, squares minimization of the residual of a set of linear or written in Fortran. nonlinear equations. FNLIB—Portable special function routines (e.g., Bessel NAG—NAG Numerical Libraries is a software prod- functions, the error function, and so on). uct sold by The Numerical Algorithms Group Ltd IMSL—The IMSL (International Mathematics and (originally the Nottingham
Load more

Recommended publications
  • Computational Economics and Econometrics
    Computational Economics and Econometrics A. Ronald Gallant Penn State University c 2015 A. Ronald Gallant Course Website http://www.aronaldg.org/courses/compecon Go to website and discuss • Preassignment • Course plan (briefly now, in more detail a few slides later) • Source code • libscl • Lectures • Homework • Projects Course Objective – Intro Introduce modern methods of computation and numerical anal- ysis to enable students to solve computationally intensive prob- lems in economics, econometrics, and finance. The key concepts to be mastered are the following: • The object oriented programming style. • The use of standard data structures. • Implementation and use of a matrix class. • Implementation and use of numerical algorithms. • Practical applications. • Parallel processing. Details follow Object Oriented Programming Object oriented programming is a style of programming devel- oped to support modern computing projects. Much of the devel- opment was in the commercial sector. The features of interest to us are the following: • The computer code closely resembles the way we think about a problem. • The computer code is compartmentalized into objects that perform clearly specified tasks. Importantly, this allows one to work on one part of the code without having to remember how the other parts work: se- lective ignorance • One can use inheritance and virtual functions both to describe the project design as interfaces to its objects and to permit polymorphism. Inter- faces cause the compiler to enforce our design, relieving us of the chore. Polymorphism allows us to easily swap in and out objects so we can try different models, different algorithms, etc. • The structures used to implement objects are much more flexible than the minimalist types of non-object oriented language structures such as subroutines, functions, and static common storage.
    [Show full text]
  • Fortran Resources 1
    Fortran Resources 1 Ian D Chivers Jane Sleightholme May 7, 2021 1The original basis for this document was Mike Metcalf’s Fortran Information File. The next input came from people on comp-fortran-90. Details of how to subscribe or browse this list can be found in this document. If you have any corrections, additions, suggestions etc to make please contact us and we will endeavor to include your comments in later versions. Thanks to all the people who have contributed. Revision history The most recent version can be found at https://www.fortranplus.co.uk/fortran-information/ and the files section of the comp-fortran-90 list. https://www.jiscmail.ac.uk/cgi-bin/webadmin?A0=comp-fortran-90 • May 2021. Major update to the Intel entry. Also changes to the editors and IDE section, the graphics section, and the parallel programming section. • October 2020. Added an entry for Nvidia to the compiler section. Nvidia has integrated the PGI compiler suite into their NVIDIA HPC SDK product. Nvidia are also contributing to the LLVM Flang project. Updated the ’Additional Compiler Information’ entry in the compiler section. The Polyhedron benchmarks discuss automatic parallelisation. The fortranplus entry covers the diagnostic capability of the Cray, gfortran, Intel, Nag, Oracle and Nvidia compilers. Updated one entry and removed three others from the software tools section. Added ’Fortran Discourse’ to the e-lists section. We have also made changes to the Latex style sheet. • September 2020. Added a computer arithmetic and IEEE formats section. • June 2020. Updated the compiler entry with details of standard conformance.
    [Show full text]
  • Over-Scalapack.Pdf
    1 2 The Situation: Parallel scienti c applications are typically Overview of ScaLAPACK written from scratch, or manually adapted from sequen- tial programs Jack Dongarra using simple SPMD mo del in Fortran or C Computer Science Department UniversityofTennessee with explicit message passing primitives and Mathematical Sciences Section Which means Oak Ridge National Lab oratory similar comp onents are co ded over and over again, co des are dicult to develop and maintain http://w ww .ne tl ib. org /ut k/p e opl e/J ackDo ngarra.html debugging particularly unpleasant... dicult to reuse comp onents not really designed for long-term software solution 3 4 What should math software lo ok like? The NetSolve Client Many more p ossibilities than shared memory Virtual Software Library Multiplicityofinterfaces ! Ease of use Possible approaches { Minimum change to status quo Programming interfaces Message passing and ob ject oriented { Reusable templates { requires sophisticated user Cinterface FORTRAN interface { Problem solving environments Integrated systems a la Matlab, Mathematica use with heterogeneous platform Interactiveinterfaces { Computational server, like netlib/xnetlib f2c Non-graphic interfaces Some users want high p erf., access to details {MATLAB interface Others sacri ce sp eed to hide details { UNIX-Shell interface Not enough p enetration of libraries into hardest appli- Graphic interfaces cations on vector machines { TK/TCL interface Need to lo ok at applications and rethink mathematical software { HotJava Browser
    [Show full text]
  • Prospectus for the Next LAPACK and Scalapack Libraries
    Prospectus for the Next LAPACK and ScaLAPACK Libraries James Demmel1, Jack Dongarra23, Beresford Parlett1, William Kahan1, Ming Gu1, David Bindel1, Yozo Hida1, Xiaoye Li1, Osni Marques1, E. Jason Riedy1, Christof Voemel1, Julien Langou2, Piotr Luszczek2, Jakub Kurzak2, Alfredo Buttari2, Julie Langou2, Stanimire Tomov2 1 University of California, Berkeley CA 94720, USA, 2 University of Tennessee, Knoxville TN 37996, USA, 3 Oak Ridge National Laboratory, Oak Ridge, TN 37831, USA, 1 Introduction Dense linear algebra (DLA) forms the core of many scienti¯c computing appli- cations. Consequently, there is continuous interest and demand for the devel- opment of increasingly better algorithms in the ¯eld. Here 'better' has a broad meaning, and includes improved reliability, accuracy, robustness, ease of use, and most importantly new or improved algorithms that would more e±ciently use the available computational resources to speed up the computation. The rapidly evolving high end computing systems and the close dependence of DLA algo- rithms on the computational environment is what makes the ¯eld particularly dynamic. A typical example of the importance and impact of this dependence is the development of LAPACK [1] (and later ScaLAPACK [2]) as a successor to the well known and formerly widely used LINPACK [3] and EISPACK [3] libraries. Both LINPACK and EISPACK were based, and their e±ciency depended, on optimized Level 1 BLAS [4]. Hardware development trends though, and in par- ticular an increasing Processor-to-Memory speed gap of approximately 50% per year, started to increasingly show the ine±ciency of Level 1 BLAS vs Level 2 and 3 BLAS, which prompted e®orts to reorganize DLA algorithms to use block matrix operations in their innermost loops.
    [Show full text]
  • Jack Dongarra: Supercomputing Expert and Mathematical Software Specialist
    Biographies Jack Dongarra: Supercomputing Expert and Mathematical Software Specialist Thomas Haigh University of Wisconsin Editor: Thomas Haigh Jack J. Dongarra was born in Chicago in 1950 to a he applied for an internship at nearby Argonne National family of Sicilian immigrants. He remembers himself as Laboratory as part of a program that rewarded a small an undistinguished student during his time at a local group of undergraduates with college credit. Dongarra Catholic elementary school, burdened by undiagnosed credits his success here, against strong competition from dyslexia.Onlylater,inhighschool,didhebeginto students attending far more prestigious schools, to Leff’s connect material from science classes with his love of friendship with Jim Pool who was then associate director taking machines apart and tinkering with them. Inspired of the Applied Mathematics Division at Argonne.2 by his science teacher, he attended Chicago State University and majored in mathematics, thinking that EISPACK this would combine well with education courses to equip Dongarra was supervised by Brian Smith, a young him for a high school teaching career. The first person in researcher whose primary concern at the time was the his family to go to college, he lived at home and worked lab’s EISPACK project. Although budget cuts forced Pool to in a pizza restaurant to cover the cost of his education.1 make substantial layoffs during his time as acting In 1972, during his senior year, a series of chance director of the Applied Mathematics Division in 1970– events reshaped Dongarra’s planned career. On the 1971, he had made a special effort to find funds to suggestion of Harvey Leff, one of his physics professors, protect the project and hire Smith.
    [Show full text]
  • Totalview: Debugging from Desktop to Supercomputer
    ATPESC 2017 TotalView: Debugging from Desktop to Supercomputer Peter Thompson Principal Software Support Engineer August 8, 2017 © 2017 Rogue Wave Software, Inc. All Rights Reserved. 1 Some thoughts on debugging • As soon as we started programming, we found out to our surprise that it wasn’t as easy to get programs right as we had thought. Debugging had to be discovered. I can remember the exact instant when I realized that a large part of my life from then on was going to be spent in finding mistakes in my own programs. – Maurice Wilkes • Debugging is twice as hard as writing the code in the first place. Therefore, if you write the code as cleverly as possible, you are, by definition, not smart enough to debug it. – Brian W. Kernigan • Sometimes it pays to stay in bed on Monday, rather than spending the rest of the week debugging Monday’s code. – Dan Saloman © 2017 Rogue Wave Software, Inc. All Rights Reserved. 2 Rogue Wave’s Debugging Tool TotalView for HPC • Source code debugger for C/C++/Fortran – Visibility into applications – Control over applications • Scalability • Usability • Support for HPC platforms and languages © 2017 Rogue Wave Software, Inc. All Rights Reserved. 3 TotalView Overview © 2017 Rogue Wave Software, Inc. All Rights Reserved. 4 TotalView Origins Mid-1980’s Bolt, Berenak, and Newman (BBN) Butterfly Machine An early ‘Massively Parallel’ computer © 2017 Rogue Wave Software, Inc. All Rights Reserved. 5 How do you debug a Butterfly? • TotalView project was developed as a solution for this environment – Able to debug multiple processes and threads – Point and click interface – Multiple and Mixed Language Support • Core development group has been there from the beginning and have been/are involved in defining MPI interfaces, DWARF, and lately OMPD (Open MP debugging interface) © 2017 Rogue Wave Software, Inc.
    [Show full text]
  • C++ API for BLAS and LAPACK
    2 C++ API for BLAS and LAPACK Mark Gates ICL1 Piotr Luszczek ICL Ahmad Abdelfattah ICL Jakub Kurzak ICL Jack Dongarra ICL Konstantin Arturov Intel2 Cris Cecka NVIDIA3 Chip Freitag AMD4 1Innovative Computing Laboratory 2Intel Corporation 3NVIDIA Corporation 4Advanced Micro Devices, Inc. February 21, 2018 This research was supported by the Exascale Computing Project (17-SC-20-SC), a collaborative effort of two U.S. Department of Energy organizations (Office of Science and the National Nuclear Security Administration) responsible for the planning and preparation of a capable exascale ecosystem, including software, applications, hardware, advanced system engineering and early testbed platforms, in support of the nation's exascale computing imperative. Revision Notes 06-2017 first publication 02-2018 • copy editing, • new cover. 03-2018 • adding a section about GPU support, • adding Ahmad Abdelfattah as an author. @techreport{gates2017cpp, author={Gates, Mark and Luszczek, Piotr and Abdelfattah, Ahmad and Kurzak, Jakub and Dongarra, Jack and Arturov, Konstantin and Cecka, Cris and Freitag, Chip}, title={{SLATE} Working Note 2: C++ {API} for {BLAS} and {LAPACK}}, institution={Innovative Computing Laboratory, University of Tennessee}, year={2017}, month={June}, number={ICL-UT-17-03}, note={revision 03-2018} } i Contents 1 Introduction and Motivation1 2 Standards and Trends2 2.1 Programming Language Fortran............................ 2 2.1.1 FORTRAN 77.................................. 2 2.1.2 BLAST XBLAS................................. 4 2.1.3 Fortran 90.................................... 4 2.2 Programming Language C................................ 5 2.2.1 Netlib CBLAS.................................. 5 2.2.2 Intel MKL CBLAS................................ 6 2.2.3 Netlib lapack cwrapper............................. 6 2.2.4 LAPACKE.................................... 7 2.2.5 Next-Generation BLAS: \BLAS G2".....................
    [Show full text]
  • IMSL C Function Catalog 2020.0
    IMSL® C Numerical Library Function Catalog Version 2020.0 1 At the heart of the IMSL C Numerical Library is a comprehensive set of pre-built mathematical and statistical analysis functions that developers can embed directly into their applications. Available for a wide range of computing platforms, the robust, scalable, portable and high performing IMSL analytics allow developers to focus on their domain of expertise and reduce development time. 2 IMSL C Numerical Library v2020.0 Function Catalog COST-EFFECTIVENESS AND VALUE including transforms, convolutions, splines, The IMSL C Numerical Library significantly shortens quadrature, and more. application time to market and promotes standardization. Descriptive function names and EMBEDDABILITY variable argument lists have been implemented to Development is made easier because library code simplify calling sequences. Using the IMSL C Library readily embeds into application code, with no reduces costs associated with the design, additional infrastructure such as app/management development, documentation, testing and consoles, servers, or programming environments maintenance of applications. Robust, scalable, needed. portable, and high performing analytics with IMSL Wrappers complicate development by requiring the forms the foundation for inherently reliable developer to access external compilers and pass applications. arrays or user-defined data types to ensure compatibility between the different languages. The A RICH SET OF PREDICTIVE ANALYTICS IMSL C Library allows developers to write, build, FUNCTIONS compile and debug code in a single environment, and The library includes a comprehensive set of functions to easily embed analytic functions in applications and for machine learning, data mining, prediction, and databases. classification, including: Time series models such as ARIMA, RELIABILITY GARCH, and VARMA vector auto- 100% pure C code maximizes robustness.
    [Show full text]
  • Small-Signal Device and Circuit Simulation
    Die approbierte Originalversion dieser Dissertation ist an der Hauptbibliothek der Technischen Universität Wien aufgestellt (http://www.ub.tuwien.ac.at). The approved original version of this thesis is available at the main library of the Vienna University of Technology (http://www.ub.tuwien.ac.at/englweb/). DISSERTATION Small-Signal Device and Circuit Simulation ausgeführt zum Zwecke der Erlangung des akademischen Grades eines Doktors der technischen Wissenschaften eingereicht an der Technischen Universität Wien Fakultät für Elektrotechnik und Informationstechnik von STEPHAN WAGNER Gauermanngasse 3 A-3003 Gablitz, Österreich Matr. Nr. 9525168 I geboren am 15. Oktober 1976 in Wien Wien, im April 2005 Kurzfassung Die Entwicklung der letzten Jahrzehnte hat gezeigt, dass die numerische Bauelementsimulation einen wichtigen Beitrag zur Charakterisierung von Halbleiterstrukturen leistet. Der wirtschaftliche Vorteil des Einsatzes von modernen Simulationsprogrammen ist enorm, da Resultate von aufwendigen und teuren Ex- perimenten durch numerische Berechnungen vorausgesagt werden können. Darüberhinaus können diese Resultate für die Optimierung der Strukturen herangezogen werden, wodurch nur die vielversprechend- sten Varianten hinsichtlich zuvor bestimmter Kenngrößen tatsächlich hergestellt werden brauchen. Neben diesen Vorteilen führte auch die breite Verfügbarkeit und großzügigere Speicherausstattung eines durch- schnittlichen Computers heutzutage zu einer stark vermehrten Anwendung dieser Simulationsprogramme. Um die Voraussetzungen für die
    [Show full text]
  • Evolving Software Repositories
    1 Evolving Software Rep ositories http://www.netli b.org/utk/pro ject s/esr/ Jack Dongarra UniversityofTennessee and Oak Ridge National Lab oratory Ron Boisvert National Institute of Standards and Technology Eric Grosse AT&T Bell Lab oratories 2 Pro ject Fo cus Areas NHSE Overview Resource Cataloging and Distribution System RCDS Safe execution environments for mobile co de Application-l evel and content-oriented to ols Rep ository interop erabili ty Distributed, semantic-based searching 3 NHSE National HPCC Software Exchange NASA plus other agencies funded CRPC pro ject Center for ResearchonParallel Computation CRPC { Argonne National Lab oratory { California Institute of Technology { Rice University { Syracuse University { UniversityofTennessee Uniform interface to distributed HPCC software rep ositories Facilitation of cross-agency and interdisciplinary software reuse Material from ASTA, HPCS, and I ITA comp onents of the HPCC program http://www.netlib.org/nhse/ 4 Goals: Capture, preserve and makeavailable all software and software- related artifacts pro duced by the federal HPCC program. Soft- ware related artifacts include algorithms, sp eci cations, designs, do cumentation, rep ort, ... Promote formation, growth, and interop eration of discipline-oriented rep ositories that organize, evaluate, and add value to individual contributions. Employ and develop where necessary state-of-the-art technologies for assisting users in nding, understanding, and using HPCC software and technologies. 5 Bene ts: 1. Faster development of high-quality software so that scientists can sp end less time writing and debugging programs and more time on research problems. 2. Less duplication of software development e ort by sharing of soft- ware mo dules.
    [Show full text]
  • IMSL™ Fortran Numerical Library V5.0 Now Available with Absoft Pro Fortran V9.0 for Mac OS X
    FOR IMMEDIATE RELEASE IMSL™ Fortran Numerical Library v5.0 Now Available with Absoft Pro Fortran v9.0 for Mac OS X World renowned mathematical & statistical libraries now bundled with new Absoft Pro Fortran Compiler v9.0 on Apple’s latest OS X Rochester Hills, MI (September 21, 2004) – Absoft Corporation has announced immediate availability of Visual Numerics’ IMSL™ Fortran Numerical Library version 5.0 now bundled with its high performance Pro Fortran Compiler Suite version 9.0 for Macintosh OS X. The IMSL™ Fortran Numerical Library is the gold standard mathematical and statistical library for Fortran programmers developing high performance computing applications. The IMSL Fortran Numerical Library saves development time by eliminating the need to write code from scratch. Using only two or three of the IMSL Fortran Library’s mathematical or statistical routines will more than pay for the product in time-savings alone. The complete IMSL Fortran Numerical Library, is now available to Absoft users on both Windows and Macintosh OS X platforms. The IMSL Fortran Numerical Library, version 5.0, includes new powerful and flexible interface modules for all applicable routines. These new interface modules allow users to utilize the fast, convenient optional arguments of the modern Fortran syntax for 100% of the relevant algorithms in the library, allowing for greater control and faster, simpler development of code. The IMSL Fortran Numerical Library provides full depth and control via optional arguments for experienced programmers, reduces development efforts by checking data-type matches and array sizing at compile time, and gives developers a simple and flexible interface to the library routines as well as speeds programming and simplifies documentation.
    [Show full text]
  • IBM Departmental Supercomputing Solutions
    Supercomputing made affordable for HPC Departmental users IBM Departmental Supercomputing Solutions Now, with the recent introduction of densely packaged rack-optimized servers and blades along with advances in software technology that make it easier to manage and exploit resources, it is possible to extend the benefits of clustered computing to individuals and small departments. Affordable Supercomputing for HPC Highlights With the IBM Departmental Departmental Users Supercomputing Solutions, barriers to Scientists, engineers and researchers deploying clustered servers—high ■ Complete High Performance have long chosen clustered servers price, complexity, floor-space require- Computing (HPC) solutions with as the preferred solution for large, ments, power consumption levels— pre-tested, densely packaged complex and computational- and have been overcome. Clients with clustered servers deliver high data-intensive problems. Clusters smaller budgets and staff but with performance at a modest price have also proven to be a cost- challenging problems to solve can effective method for managing mixed ■ Includes innovative software tools now leverage the same supercom- HPC workloads. It is common today making cluster systems manage- puting technology used by large for clusters to be used by large cor- ment easy while reducing the time organizations and prestigious labs. porations, universities and govern- it takes to use the solution ment labs to solve problems in life productively IBM Departmental Supercomputing sciences, petroleum exploration,
    [Show full text]