- Analog
- A Web log analysis program. The features include a fast response
time, ease of installation and use, flexibility (i.e. over 180 options
producing 17 different reports), the capability of creating output in
7 different languages in 3 output formats, the production of aesthetically
pleasing output that complies with
HTML specs, understanding several
different log formats (e.g. common log format, old-style NCSA format,
NCSA/Apache referrer log format, etc.), and much.
It's freely available under the conditions of a license and
is available for any UNIX, Mac or VMS machine and almost all PCs.
The documentation is contained within a large README file available
in ASCII and HTML formats.
[http://www.statslab.cam.ac.uk/~sret1/analog/]
- ANALGWST
- A set of programs to calculate analytical solutions for 1-, 2-, and
3-D solute transport in groundwater systems with uniform flow.
Analytical solutions are useful for predicting the fate of solutes
in ground water.
The individual programs comprising the package are:
- finite, for 1-D solute transport in a finite system;
- seminf, for 1-D solute transport in a semi-infinite system;
- point2, for 2-D solute transport in an infinite system with
a continuous point source;
- stripf, for 2-D solute transport in a finite-width system
with a finite-width solute source;
- stripi, for 2-D solute transport in an infinite-width system
with a finite-width solute source;
- gauss, for 2-D solute transport in an infinite-width system
with solute source having a Gaussian concentration distribution;
- point3, for 3-D solute transport in an infinite system with
a continuous point source;
- point3_mod, a modified point3 version;
- patchf, for 3-D solute transport in a finite-width and
finite-height system with a finite-width and finite-height source; and
- patchi, for 3-D solute transport in an infinite-width and
infinite-height system with a finite-width and finite-height source.
A source code distribution of ANALGWST for UNIX platforms is
available.
The primary documentation is contained within
Wexler (1992) and Wexler (1992).
This is part of the USGS
Water Resources Applications Software
collection.
[http://water.usgs.gov/software/analgwst.html]
- ANALYZE
- A package designed to provide computer assistance for
analyzing linear programs and their solutions. It is presumed
that a linear program has already been formulated and an instance
has been generated with some language. It has three levels of
use: (1) providing a convenient interactive query to navigate
through a linear program with or without a solution already having
been obtained from some solver; (2) providing procedures to assist
analysis in a variety of ways, e.g. answers to standard sensitivity
questions, etc.; and (3) providing an artificially intelligent
environment with results automatically translated into English.
An executable binary version of ANALYZE is available for both
DOS and Linux platforms. The official manual is a pricey
beastie, more information about which can be found at the
IMPS Software
site. You might want to also check out the
MODLER and
RANDMOD packages available
at the same site.
A quick summary of the ins and outs of optimization can be found at the
NEOS Guide Optimization Tree
.
See Greenberg (1993).
[http://www-math.cudenver.edu/~hgreenbe/consortium/
softget.html]
- Angel
- A tool for monitoring services on a network, Angel is a
Perl program that runs at periodic
intervals and calls various Perl subprograms that do the
actual testing. It then generates an HTML table containing
information on network status.
The features include:
- centralized administration, i.e. only one main file needs to be
modified;
- easy customization and extension via a plug-in concept;
- monitoring of results via the Web;
- support for hte LEDSign Java applet;
- support for Javascript; and
- ssh support.
A source code distribution is available.
[http://www.ism.com.br/~paganini//angel/]
[http://ibm-0.MPA-Garching.MPG.DE/angel/
]
- anim
- A system for algorithm animation developed by Jon Bentley and
Brian Kernighan.
A description of an animation is created in a special
language and the result can either be shown as a movie on X11 or
used to create still shots.
See Bentley and Kernighan (1991) which is also available as
Computing Science Tech Report 132 in the research/cstr
directory at the same site.
[http://www.netlib.org/research/]
- ANNIE
- An interactive hydrologic analysis and data management package designed
to help users interactively store, retrieve, list, plot, echeck, and
update spatial, parametric, and time-series data for hydrologic
models and analyses.
The data are stored in direct access files in Watershed Data
Management (WDM) format, a format used by many software packages
developed by the USGS and EPA.
WDM files are binary, direct-access files organized into data sets,
with each set containing a specific type of data, e.g. streamflow
at a specific site or air temperature at a weather station.
Each data set also contains attributes that describe the data, i.e.
metadata.
Data sets in other formats (e.g. flat ASCII files) can be converted
to WDM format using the
IOWDM program.
A source code distribution of ANNIE for UNIX platforms is
available.
The primary documentation is contained within
Flynn et al. (1995).
This is part of the USGS
Water Resources Applications Software
collection.
[http://water.usgs.gov/software/annie.html]
- ANNLIB
- See SPRANNLIB.
- ANRAY
- A seismological package that can be used for the computation of rays,
travel times, ray amplitudes, and ray synthetic seismograms in 3-D
laterally varying structures containing isotropic and/or anisotropic
layers. Synthetic seismograms can be constructed at receivers
distributed regularly or irregularly along the surface, at interfaces,
or on vertical profiles.
The nine programs comprising ANRAY are:
- ANRAY, the basic program designed for ray, travel time, and
ray amplitude computations in one of two modes (initial-value
or two-point ray tracing), with polarization vectors, geometrical
spreading and reflection, transmission and conversion coefficients
evaluated along the rays;
- ANRAYPL, for plotting horizontal and vertical ray diagrams as
well as time-distance and amplitude-distance curves of individual
elementary waves computed by ANRAY;
- SYNTAN, used for the computation of ray synthetic seismograms
using Gabor wavelets as the source-time function;
- FRESAN and SYNFAN, used for the frequency domain
computation of ray synthetic seismograms, with the former computing
the frequency response and the latter multiplying it by the spectrum
of the considered source-time function;
- SEISPLOT, for plotting synthetic seismograms generated
by SYNTAN;
- BPLOT, for plotting those generated by SYNTAN;
- POLARPLOT, for plotting polar motion diagrams for results
from either; and
- VELPLOT, for plotting plane sections of slowness, phase
velocity and group velocity surfaces from the results of
ANRAY.
A source code distribution of the ANRAY programs is available.
It is written in Fortran 77 with the graphics routines throughout
making calls to standard CALCOMP routines.
Documentation is contained mainly within the source code files
themselves as well as in some technical reports.
[http://seis.karlov.mff.cuni.cz/consort/main.htm]
- ANTLR
- A language tool which provides a framework for constructing
recognizers, compilers, and translators from grammatical
descriptions containing C,
C++, or Java actions.
This was formerly knowns as the PCCTS, with PCCTS 1.33 having
consisted of a lexical analyzer generator (DLG), a parser
generator (ANTLR), and a tree parser generator (SORCERER).
ANTLR 2.00 is a complete rewrite of PCCTS 1.33 in Java.
It embodies all three of the previous PCCTS tools although it
generalizes the notion of scanning, parsing, and tree walking
into the simpler idea of applying grammatical structure to an
input stream containing characters, tokens, or tree nodes.
ANTLR 2.00 generates only Java but there are plans to extend it
to generate other languages such as C++.
See Parr (1997).
[http://www.ANTLR.org/]
[http://www.mcs.net/~tmoog/pccts.html
]
[http://www.empathy.com/pccts/index.html
]
- AOLserver
- A web server with a multithreaded architecture.
AOLserver uses threads to achieve
fast response times and the multiple, simultaneous servicing
of connections.
An nsthread platform independent C API for multithreaded
programming includes functions to create and wait for new
threads, thread local storage for maintaining per-thread data,
and mutex, critical section, semaphore and event objects to
protect shared resources.
The HTTP protocol is implemented over underlying network protocols
through the use of plug-in communcations drivers.
These include:
nssock, for implementing HTTP over TCP/IP sockets;
nsssl, for implementing HTTP over SSL
TCP/IP sockets; and
nsfile, which implements HTTP using regular file system
input and output files for black box testing purposes.
A full text engine is also included as is a platform independent
interface to SQL relational databases.
AOLserver includes a
complete C API which allows the user to write:
custom request functions to handle HTTP
requests to a URL,
custom URL-to-file translation routines to convert an HTTP URL
to a file in a local filesystem,
database drivers to interface the nsdb module to an external DBMS,
a communications driver so HTTP can be implemented over new
underlying protocols,
request trace functions which run after each HTTP request, and
a scheduled procedure which runs at regular intervals to implement, e.g.
a statistics gathering system.
The Tcl scripting language is included
as an integral part of the architecture.
This interface is also multithreaded to allow more than one Tcl
script to simultaneously operate.
The interface can be extended to include custom Tcl commands,
includes a module with commands to access open databases,
provides a command for opening TCP/IP sockets to implement quick
HTTP transactions, and includes useful commands for accessing
HTML form data.
Binary distributions of AOLserver are available for several
platforms include Linux Intel.
Extensive documentation is separately available in
several formats.
[http://www.aolserver.com/server/]
- Apache
- A PAtCHy server is
an HTTP
server that is a plug-in replacement for NCSA 1.3 and much more.
In addition to fixing bugs and security holes seen in the latter, is
more efficient and faster, offers better compliance with existing
HTTP specs, and implements additional features. These include
DBM databases for authentication, customized responses to errors
and problems, multiple directory index directives, unlimited
numbers of alias and redirect directives, content negotiation,
multi-homed servers, and more.
A useful feature of Apache is modules, i.e. plug-in programs
that use a standard API to add additional features to the server.
Modules included in the standard distribution include those for
performing actions based on assigned MIME types,
basic user authentication, controlling per directory access,
CGI execution,
dynamically loading Apache modules,
a configurable log,
user authenticatin using a DB format database,
controlling filesystem mapping of user directories,
generating directory indexes on the fly,
cookie generation and tracking,
content negotiation of MIME types,
proxy support,
providing server status information,
server side includes,
handling imagemap files,
handling MIME types,
passing environment variables to CGI/SSI scripts,
common log format logging,
directory aliasing and redirects,
a log referer,
CERN meta file emulation, and more.
Features new to Apache version 1.2 (1/98) include:
- conditional compliance with the HPPT/1.1 proposed standard for
greater performance and efficieny when transferring files;
- extended server side includes (XSSI) which are directives that
allow users to better create HTML pages;
- file-based and regex-enabled directive sections which allow
directives to be enabled based on full filename rather than just
directory and URL;
- browser-based environment variables that, combined with XSSI,
allow the writing of browser-based conditional HTML documents;
- the execution of CGI scripts as users other than the server
user via setUID;
- a URL rewriting mode which provides powerful URL mapping using
regular expressions;
- enhanced, configurable logging that can log more details about
transactions as well as open more than one log file at once;
- user tracking (i.e. cookies) revision that makes it possible to
disable the generation of cookies;
- VirtualHost enhancements that allow more than one IP address
or hostname which lets a single vhost handle requests for multiple
IPs or hostnames;
- resource limits for CGI scripts;
- a CGI debugging environment which allows the setting up of a log
that records all input and output to failed CGI scripts;
- the mod_headers module can be used to set custom headers
in the HTTP response;
- better compatibility with NCSA 1.5; and
- an improved FTP, HTTP, and CONNECT mode SSL proxy with improved
FTP proxy supporting PASV mode, a ProxyBlock directive for excluding
sites to proxy, and a NoCache directive for disabling proxy cahing.
There are additional contributed Apache modules available for
mSQL authentication,
setting user/group ID for CGI execution,
faking basic authentication using cookies,
basic authentication using system accounts,
allowing or denying access to user/domain pairs,
authenticating users from an LDAP directory,
an embedded Perl interpreter,
disallowing serving pages based on UID/GID,
viewing an FTP archive using WWW,
implementing LDAP authentication and access rules,
enabling direct execution of Java applets as CGI,
using the heitml package,
using the PHP/FI package,
determing MIME type from file contents,
using an embedded Python interpreter,
Kerberos authentication,
limiting bandwidth based on number of connections,
using the NeoWebScript package,
and many more.
A source code distribution of Apache is available as are
binaries for just about every platform made.
It is written in C and can be easily compiled on most
platforms.
Documentation is available online, although the most complete
available guide is
Laurie and Laurie (1997).
[http://www.apache.org/]
- mod-perl
- A package which makes it possible to write Apache
modules entirely in Perl.
A persistent interpreter embedded in the server avoids the overhead
of starting an external interpreter and the penalty of Perl
startup time.
This is accomplished by linking the Perl runtime library into the
server and providing an object-oriented Perl interface to the server's
C language API.
Apache modules written in mod-perl can do just about anything that
modules written in C can do, with nearly equivalent speed.
This is a Perl module.
[http://perl.apache.org/]
- apfloat
- A high performance arbitrary precision arithmetic package.
Multiplications are performed using fast number theoretic
transforms with three different moduli and the Chinese Remainder
Theorem for optimal memory usage, maximum speed and no roundoff
errors. Calculations involving billions of digits can be performed
using apfloat. Some of the more recent features added to the
package include support for arbitrary bases, complex number
arithmetic, floor and ceiling functions, and order(n log n)
iterations for log, exponential, trig and hyperbolic functions
and their inverses.
Apfloat is written in C++ and will compile using most C++ compilers,
although gcc is recommended. The package also includes assembler
optimizations for 486 and Pentium processors for maximum performance
on PCs. The package comprises a common source file for all
systems plus an additional file depending on one's specific
system, e.g. Linux, Alpha, general 32-bit UNIX, etc. The
documentation is contained in a 30+ page document in
PostScript format.
[http://www.hut.fi/~mtommila/apfloat/]
- APL
- A Programming Language (or Array
Programming Language) was created at IBM in
the 1960s by Ken Iverson and others.
Its main purpose was to serve as a powerful executable notation
for mathematical algorithms, and it is probably best known for
its use of several non-ASCII symbols including some Greek letters.
It is a dynamically typed, interactive, array-oriented language
with dynamic scoping in which all expressions are evaluated from
right to left.
See Grey (1973),
LePage (1978),
Polivka and Pakin (1975), and
Rose and Schick (1980).
[http://www.chilton.com/~jimw/]
[ftp://watserv1.uwaterloo.ca/languages/apl/Welcome.html
]
- aplc
- An APL to C translator. Check the files called status* in the
directory for the status of this project.
[ftp://csi.jpl.nasa.gov/pub/apl/]
[ftp://ftp.cs.orst.edu/pub/budd/
]
- APL-11
- An APL interpreter for the UNIX operating system.
It is written in C and a source code implementation is
available.
[ftp://watserv1.uwaterloo.ca/languages/apl/apl-11/]
- A++/P++
- A++ is a C++ array
class for numerical computation designed to work with structured
grid computations, including work on overlapping grids and adaptive
mesh refinement. P++ is the parallel version of the serial
class A++. A++/P++ was designed to simplify the development of
numerical software, specifically to allow an application developed
in the serial environment to be run on parallel machines with little
or no additional effort. It is also intended as a partial solution
to a growing crisis in the development of large numerical codes that
are required to run on many different serial and complex parallel
architectures.
The A++/P++ distribution (35 Mb uncompressed) includes the source
code and the documentation. Requirements for compilation and
installation include a C++ compiler
(g++ will do) and a C and/or
a Fortran compiler.
The graphics visualization facilities of
A++/P++ additionally require the
Plotmtv software.
The use of P++ requires a communication library. It presently
works with both
MPI and
PVM. This is affiliated with
the related POOMA Project
,
with which it is eventually planned to merge
A++/P++.
[http://www.c3.lanl.gov/~dquinlan/A++P++.html]
- AppGEN
- A high level fourth generation language (4GL) and application
generator for producing WWW based applications which are
typically used over the Internet or within an intranet.
AppGEN applications are implemented as C scripts conforming
to the CGI standard.
AppGEN consists of several programs:
- defgen, which produces a basic template application from a logical
data structure with the applications capable of adding,
updating, deleting and searching for records within the database
whilst automatically maintaining referential integrity;
- appgen, the AppGEN compiler which compiles the appgen source
code into CGI executable C source and HTML documents ready for
deployment; and
- dbf2sql, a utility for converting dBase III compatible
i .dbf files in executable SQL scripts which enables the data stored in
most DOS/Windows based packages to be migrated to a SQL server
such as PostgreSQL.
The distribution also includes a collection of HTML documents,
GIF files and Java applets which are used at runtime by the
system.
The current distribution includes Linux ELF binaries for
the three programs, the ancillary programs and files, and
the full source code.
The use of AppGEN requires PostgreSQL,
a CGI compatible web server, and an ANSI C compiler.
[http://www.man.ac.uk/~whaley/ag/appgen.html]
- Apple II+ Emulator
- An emulator for the Apple II+
computer written especially for Linux which takes
advantage of the SVGA library distribution.
It is written partly in assembler and partly in C for
speed, and runs about twice as fast as a real II+ on
a 486 DX-50 system.
[http://geta.life.uiuc.edu/~badger/files/]
- April
- A process-oriented language for implementing intelligent applications
on a network as well as a platform to execute applications.
It includes several features needed for network computing including
a structure for uniquely identifying an agent on a global network,
a method for defining mobile codes, a data structure which enables
easy exchange of messages between agents, and security features
to protect against harmful codes.
The April distribution includes:
april, a run-time system;
ac, the April compiler;
apcommserver, a communications server program;
aplist, which lists nameserver contents;
apdebugger, a simple debugger server;
apshell, a simple remote fork server program; and
apdump, a utility program to display the contents of encoded files.
Binary distributions are available for some platforms including
Linux Intel.
Documentation is included in the distribution.
[http://www.fujitsu.co.jp/hypertext/Products/Software/April/Eindex.html]
- Ara
- A platform for the portable and secure execution of mobile
agents in heterogeneous networks, where mobile agents are
programs with the ability to change their host machine during
execution while preserving their internal state.
The aim of the Ara project is to provide full mobile
agent functionality while retaining as much as possible of
established programming models and languages.
The application focus of Ara is on weakly connected and high
volume systems such as wireless or intermittently connected
computers or globally distributed large data bases, which
seem particularly well suited for such applications.
The system architecture consists of a core an several processes
where agents are executed as processes, with the complete
system running as a single application process on top of an
unmodified host operating system.
The agents are executed within an interpreter for their
respective languages, and Ara defines an interface within which
interpreters for established languages can be used.
Thus far (4/97) the languages which have been adapted to
Ara are Tcl and
C/C++ (by means of precompilation to MACE, an interpretable
byte code).
The Ara package is available as source code.
It has been compiled and tested on Sun Solaris and SunOS and
Linux Intel platforms.
The package is documented in several reports and guides included
in the distribution in both PostScript and HTML format.
[http://www.uni-kl.de/AG-Nehmer/Projekte/Ara/index.html]
- Arachne
- A toolkit for component-based computing developed as part of several
projects concerned advanced health care software systems.
Arachne is a partial substrate for the construction and integration
of modular software capabilities intended to address long-term
issues in the creation of component-based software markets.
It consists of a collection of useful capabilities including:
portability layers for cooperative multithreading, GUI development,
and other parts of a software environment;
a full CORBA implementation that is nearly
compliant with the CORBA-2 standard;
a partial implementation of the Common Object Services (COS) associated
with the CORBA standard;
a range of component interface standards for various aspects of the
application development enviroment;
a class library for application development and a range of GUI and
data management component implementations; and
a set of associated utilities such as tools for the textual representation
of object instance data and an extensible arithmetic and boolean
evaluation engine.
The current (1/98) Arachne release consists of several subsystems including:
- CU, basic cross-platform utilities for C/C++ development
including a cooperative multithreading API;
- OD, textual object instance file format utilities;
- DB, database integration utilities and low-level database support
(i.e. the Berkeley DB engine);
- CG, a cross-platform GUI development layer;
- GEN, a full CORBA implementation;
- COS, a subset of the OMG standard COS specification;
- AR, abstract object interface classes (in IDL) and basic support
for CORBA class development;
- AOS, Arachne Object Services including an implementation of the
Externalization service based on the Berkeley DB engine; and
- AC, the Arachne class library.
Tools that are not part of the Arachne core include:
- EV, a simple extensible arithmetic and boolean evaluation engine;
- SK, a BSD sockets layer for Macintosh with support
for multithreading;
- XP-RPC, a port of Sun ONC to Macintosh based on SK; and
- WW, i.e. WebWorks, an incomplete Arachne application authoring
environment.
Additional tools are under development.
Source code distributions of the Arachne components are available
for Linux, Windows NT/95, HP/UX, SunOS and Macintosh platforms.
All are freely available for non-commercial use.
Documentation is scattered about in several file formats.
[http://dsg.harvard.edu/public/arachne/]
- arcem
- An emulator for the
Acorn Archimedes A3xx to A4xx series computer.
[ftp://ftp.compsoc.man.ac.uk:/pub/arcem/]
- ARCH
- An object-oriented library of tools for parallel programming
on machines using the MPI
communication library. It offers a set of flexible programming
constructs for parallel software development for asynchronous
and loosely synchronous system programming, creating the illusion
of shared memory on distributed memory machines. ARCH allows
the distribution of arrays as well as user-defined data
structures such as pointers to remote data. This should install
and run on machines running MPI.
[ftp://ftp.tc.cornell.edu/pub/ARCH/]
- Arena
- A general-purpose Web browser built on top of the multithreaded
version of the
W3C Reference Library.
It was originally created by the W3 Consortium as a test bed
for advanced HTML specification features, but was taken over
by Yggdrasil in early 1997 when W3 began using the
Amaya package as a testbed.
The current (7/97) version of Arena supports HTML 3.0 (i.e. the
HTML 3.2 standard preprocessor which includes the MATH tag,
tables, forms, etc.), cascading style sheets (CSS), style
sheet editing, MIME, direct access to
WAIS engines, HTTP 1.1,
HTML editing via an external editor, external client communication
(i.e. allowing other applications to know what URL is being
displayed by Arena), and the
PNG,
JPEG, and
GIF graphics formats.
As of 7/97 Yggdrasil is providing weekly developer source
code releases until it decides that it's stable enough to
start providing binary releases. It has updated Arena to
use the latest W3C Reference Library and plans to make several
more improvements.
[http://www.yggdrasil.com/Products/Arena/]
[http://www.w3.org/pub/WWW/Arena/
]
- ARfit
- A Matlab package for the estimation and spectral decomposition
of multivariate autoregressive models.
ARfit includes routines for the
estimation of parameters and confidence regions for multivariate
autoregressive (AR) processes,
diagnostic checking of fitted models, and
spectral decomposition of AR models.
The package is documented in a pair of journal papers available
online in PostScript format.
[http://solon.cma.univie.ac.at/~neum/software/arfit/]
- ARIBAS
- An interactive interpreter for big integer arithmetic and
multi-precision floating point arithmetic with a
Pascal/Modula-like
syntax. It has several built-in functions for algorithmic number
theory, e.g. gcd, Jacobi symbol, Rabin probabilitistic prime test,
Morrison-Brillhart continued fraction factorization,
and Pollard rho factorization.
A source code version of ARIBAS is available for
UNIX platforms.
It is mostly written in C, although some of the critical routines
are written in assembler for the Linux version so it will run
especially fast under Linux Intel.
[http://www.mathematik.uni-muenchen.de/~forster/sw/aribas.html]
- Arjuna
- An object-oriented programming system that provides a set of
tools for the construction of fault-tolerant distributed applications.
A prototype version written in C++ has
been designed and implemented to run on a collection of UNIX
workstations connected by a local area network. Arjuna provides nested
atomic actions (nested atomic transactions) for structuring application
programs. Atomic actions operate on objects, which are instances of
abstract data types (C++ classes),
by making use of remote procedure
calls (RPCs).
The system has been ported to various platforms including
HP workstations running HPUX; Sun 3, Sun 4, Sun Sparc and running
SunOS 4 and 5 (Solaris); RS6000 machines running AIX; and Intel
486 PCs running Linux.
The source code is available as well as a user's manual in
PostScript and
HTML formats.
[http://arjuna.ncl.ac.uk/]
- Arla
- A free client implementation of the Andrew File System (AFS).
This is planned (6/98) to be a fully functional client with all the
capabilities of the normal AFS, with management tools and a server
in the longer range plans.
[http://www.stacken.kth.se/projekt/arla/]
- ARPACK
- A collection of Fortran
77 subroutines for solving large
scale eigenvalue problems. This is designed to compute a few
eigenvalues and corresponding eigenvectors of a general NxN
matrix, and is most appropriate for large sparse or structured
matrices (where structured means that a matrix-vector product
requires order N rather than the usual order N2 floating
point operations).
ARPACK is based on an algorithmic variant of the Arnoldi
process called the implicitly started Arnoldi method (IRAM).
This reduces to a variant of the Lanczos process called
the implicitly restarted Lanczos method (IRLM) when the matrix
is symmetric.
Both variants can be viewed as a synthesis of the Arnoldi/Lanczos
proess with the implicitly shifted QR (ISQR) technique which is
suitable for large scale problems.
This software is capable of solving problems from significant
application areas and is designed to compute a few eigenvalues
with user-specified features such as those of largest real part
or largest magnitude.
Numerically accurate eigenvectors are also available on request.
The features of ARPACK include:
a reverse communication interface;
single and double precision real arithmetic versions for symmetric,
non-symmetric, standard or generalized problems;
single and double precision complex arithmetic versions for standard
or generalized problems;
routines for standard or generalized problems with banded matrices;
routines for singular value decomposition (SVD);
and example driver routines which may be used as templates
to implement numerous Shift-Invert strategies for all problem
types, data types, and precisions.
Release 2.1 (3/97) of ARPACK includes as an extension
the PARPACK library which extends the package for
use on heterogeneous clusters of workstations using either
BLACS or MPI.
ARPACK/PARPACK is a component of
the ScaLAPACK Project
.
[ftp://ftp.caam.rice.edu/pub/software/ARPACK/]
[http://www.netlib.org/linalg/
]
- ARPACK++
- An object-oriented version of ARPACK
written in C++.
ARPACK++ uses templates to reduce the work needed to establish
and solve eigenvalue problems and to simplify the structure
used to handle such problems.
It also features an interface that avoids the complication of the
reverse communication interface that characterizes the Fortran
version, has the ability to easily find interior eigenvalues and to
solve generalized problems, and a structure that minimizes the
work needed to generate an interface between it and other libraries
such as the TNT.
A source code distribution of ARPACK++ is available.
It contains a make file specifically for G++
as well as those for other compilers.
It is documented in a 200 page user's guide and reference
manual available in PostScript
format.
[http://www.caam.rice.edu/software/ARPACK/arpack++.html]
[http://www.ime.unicamp.br/~chico/arpack++/
]
- Arrow
- A mail user agent designed for
new Linux users who are used to the gee-gaws found on other,
inferior operating systems.
The features include:
- use of standard mail format;
- use of Pine's address book format;
- rearranging messages via drag and drop;
- decryption and encryption with PGP;
- drag and drop text and addresses between messages; and
- automatic saving of complete state when closing or
when the X server crashes.
The source code is available as are binaries for Linux Intel
platforms.
This was developed with the JX toolkit.
[http://www.cco.caltech.edu/~glenn/arrow/]
- artificial life
- A research discipline that studies natural life by attempting to
recreate biological phenomena from scratch within computers.
The links are to the Zooland site which contains further information
and various a-life software packages.
[http://alife.santafe.edu/~joke/zooland/]
[http://www.krl.caltech.edu/~brown/alife/zooland/
]
[http://research.de.uu.net:8080/zooland/
]
- ASCEND
- A large-scale, object-oriented mathematical modeling environment
and strongly typed mathematical modeling language.
ASCEND was primarily developed for use by chemical engineers,
although the package is domain independent and can be used in
all appropriate areas of science and engineering.
It was designed to reduce the time needed for creating, debugging, and
solving mathematical models by orders of magnitude in comparison
with languages like C++ and Fortran.
ASCEND includes a wide range of support tools for modeling, debugging,
and solving systems with upwards of tens of thousands of nonlinear
algebraic or differential equations including:
- a library of equilibrium-based unit operations for chemical
engineering;
- a user-centered structured methodology for reaching correct
problem specifications;
- automatic analysis of degrees of freedom;
- automatic initialization of variables;
- a choice of automatic scaling methods for nonlinear equations; and
- an object-oriented modeling language.
The modeling language includes such features as model construction
by object and value passing, a preprocessor that diagnoses errors and
symptoms of poor object-oriented programming style, a SELECT statement
for choosing among alternative constructions at instantiation time, and
a SWITCH statement for managing the flow of control.
Solver features include:
the selection of equations and models based on logical, integer,
symbolic configuration variables, and
the fast automatic analysis of model hierarchies to obtain
optimum ordering for fast linear factorization.
A source code version of ASCEND is available for UNIX platforms.
It can be built on several UNIX flavors including Linux and additionally
requires Tcl/Tk version 8.0p2.
It is documented in a 235 page manual available in PDF format.
[http://www.cs.cmu.edu/~ascend/]
- ASGL
- A graphics program for producing PostScript output such as scatter
plots, line plots, histograms, 2-D density plots, and
bond-and-stick plots of molecules.
A source code file is created which is interpreted by a program
called TOP which calls appropriate Fortran subroutines
to create a PostScript output file.
This is written in Fortran 77 to compile on most UNIX systems.
A user's manual is available in PostScript format.
[http://guitar.rockefeller.edu/asgl/asgl.html]
- ASHE
- See XHTML.
- asl
- A UNIX/C version of a cross assembler for a variety of
microprocessors and controllers.
The capabilities include macros, conditional assembly,
extensive listings, local symbol domains, and include files.
It supports nearly 50 different target processors from
4-bit microcontrollers to the PowerPC.
A manual is available in TeX format.
[http://sunsite.unc.edu/pub/Linux/devel/lang/assemblers/]
[ftp://zam091.zam.kfa-juelich.de/pub/msdos/assembler/
]
- ASN.1
- The Abstract Syntax Notation 1 is a
formal language for abstractly describing data exchanged between
computer systems which is very much like a type declaration
in C or C++.
It frees protocol designers to describe the
layout of messages exchanged between application programs running
in different hardware and software environments. It is called
abstract because it describes the details of the messages in a more
abstract level than bits and bytes.
Encoding rules are described which are used to transform data
specified in the ASN.1 language into a standard format which can be
decoded by any system with a decoder based on the same set of rules.
Related software packages include ASN-EZE,
ELROS,
ISODE, OSIkit, and
OSIMIS.
- ASN-EZE
- An ASN.1 to C++
compiler and library which implements the X.208 and X.209 ASN.1
syntax and encoding rules as well as the Distinguished Encoding
Rules as specified in X.509.
The distribution consists of two parts, the compiler and the
support library. The former is written in C and the latter
in C++.
The compiler, called asn_gen, takes a file of ASN.1
specifications and converts the data to C++ source code and
header files suitable for compiling and using with the ASN.1
library functions defined in asn_objects.
The source code for ASN-EZE is available and is written in
C and C++. It has been compiled with the gcc
compiler suite.
The compiler and library are documented in separate man pages as
well as in a couple of ASCII text files.
[http://ests.bbn.com/ASNSRC.html]
- ASpecT
- A strict functional language developed at the University of
Bremen in Germany. It was originally intended to provide an
implementation for a subset of algebraic specifications of
abstract datatypes and included several user-friendly features
like overloading facilities and a source-level debugger.
It also uses call-by-value evaluation and reference counting
memory management for efficiency.
Other features include subsorting, functionals, and
restricted polymorphism.
The compiler translates the functional source code into C
which can be compiled with the native C compiler to yield
fast and efficient executables.
The ASpecT system is available in binary format for
several platforms including Linux Intel, Mac, NeXT,
OS/2, Sun Solaris and SunOS, and VAX.
Documentation is included in the distributions.
The interactive graph visualization system
daVinci is written in ASpecT.
[ftp://ftp.Uni-Bremen.DE/pub/programming/languages/ASpecT]
- astkit
- The Advanced Software Technology Kit
is a set of executables and libraries provided as part of the
software accompanying Krishnamurthy (1995).
The astkit has three purposes: (1) to present a subset of the
POSIX/ANSI standard headers and interfaces
on non-compliant systems; (2) to provide a portable base of routines
that implement concepts common to all UNIX variants; and
(3) to provide a forum for modern implementations of features
present (or lacking) in the standard C libraries.
The major components of astkit include:
- ksh93, a version of the KornShell
shell programming language described in
Bolsky and Korn (1995) which is upward compatible with sh and
provides an enhanced programming environment in addition to the
major features of the BSD shell csh;
- nmake, a modern variant of the traditional make which
maintains state that records information for future runs, i.e. it records
such things as file modification times, explicit and implicit prerequisites,
action test, variable values, and target attributes;
- coshell, which executes shell blocks on lightly loaded hosts in
a local network via a server that runs as a daemon on the user's home host;
- cql, a C query language which applies C-style expressions to
flat database files;
- pax, a POSIX 1003.2 conformat replacement for
tar and cpio that handles
most UNIX archive and tape formats;
- 3d, a program that allows entry into nDFS, a multiple dimension
file system implemented as a shared library which intercepts pathname
systems calls and provides a logical namespace on top of the underlying
physical file system (and works with ksh93);
- tw, a combination of find and xargs that applies
C-style expressions on that stat structure of each file in a direcotry
hierarchy;
- proto, which converts ANSI C prototype constructs to constructs
compatible with K&R, ANSI C, and C++; and
- sort, a faster version of the traditional sort which is
a distribution sort with preconditioning of difficult keys to make them
into simple strings.
Other programs in the distribution include:
- bax, which generates backup tapes using tw and pax;
- cpp, a standalone preprocessor which is primarily intended for
C but also supports other languages;
- iffe, a command interpreter that probes the C compilation
environment for features, i.e. any file, option or symbol that
controls or is controlled by the compiler;
- mm2html, which converts mm/man input to HTML;
- html2rtf, which converts HTML input to RTF;
- troff2html, which converts troff intput to HTML;
- ss, which lists the system status for hosts on
the local network; and
- probe, which maintains tool specific information for language
processors and performs consistency and security checks to ensure
that the tool specific command generates valid information.
Over 30 additional programs and libraries are included.
Binary distributions of astkit are available for
Sun SunOS and Solaris, SGI IRIX, HP-UX, Linux Intel, and BSD Intel
platforms. Documentation includes the usual set of man pages and,
of course, the book previously mentioned.
[http://portal.research.bell-labs.com/orgs/ssr/book/reuse/]
- astronomical software
- A more specific list of software for use by the
astronomy community. In addition to the software there
is also a mailing list for those interested in porting
software packages for reducing astronomical data to Linux.
Available packages related to astronomy include:
- AIPS, a system for the calibration, analysis
and display of radio astronomy images;
- AIPS++, a C++ version
(evolved) of AIPS;
- eclipse, a set of image processing
utilities for astronomical data reduction;
- ESO-MIDAS, general tools for image
processing and data reduction with an emphasis on astronomy;
- FIGARO, a general data reduction package
aimed specifically towards optical and infrared data;
- FITS, the generic data storage format of
the astronomy community;
- FTOOLS, a collection of utility programs
to view and manipulate FITS data files;
- GILDAS, a collection of applications
oriented towards radio astronomy;
- GIPSY, an interactive system for the reduction
and display of astronomical data;
- IRAF, a general purpose system for the reduction
and analysis of astronomical data;
- MIRIAD, a toolbox for the calibration, mapping,
deconvolution and image analysis of interferometric data;
- NEMO, an extensible stellar dynamics toolbox
for analysis and visualization;
- SAOimage, a program for displaying
astronomical;
- SAORD, a collection of software including an
updated version of SAOimage;
- Skycat, a tool for image visualization and
accessing astronomy data catalogs;
- STARLINK, a comprehensive library of
astronomical software;
- STSDAS, a system for reducing and analyzing
data from the Hubble space telescope;
- WCSTools, a package of programs and libraries
for taking advantage of world coordinate system information in
astronomical image files and catalogs;
- XANADU, a system for the analysis and reduction
of data in X-ray astronomy;
- Xephem, an interactive astronomical ephemeris
program;
- xsky, an interactive sky atlas;
[http://bima.astro.umd.edu/nemo/linuxastro/]
- ASURV
- A package containing source code and documentation for
Astronomy SURVival analysis.
This implements a suite of
statistical methods for the analysis of censored data; i.e. data
which are known to lie above or below some limit. It was written
specifically to treat left-censoring arising in observational astronomy
when objects are observed but sometimes not detected due to sensitivity
limits. However, the methods can be useful to researchers in other
disciplines, as the code includes techniques that are often omitted
from commercial survival analysis packages.
ASURV computes: the maximum-likelihood Kaplan-Meier estimator; several
univariate two-sample tests (Gehan, Peto-Peto, Peto-Prentice); three
bivariate correlation coefficients (Cox regression, generalized Kendall's
tau and Spearman's rho); and three linear regressions (EM algorithm
assuming normal residuals, Buckley-James line, Schmitt line).
The program is stand-alone and does
not call any specialized library.
The package includes the source code which is written in
Fortran
and a 40+ page manual in LaTeX
format. The manual contains the
gory details about the algorithms mentioned above as well as
instructions on how to use them.
See LaValley et al. (1990),
Feigelson and Nelson (1985), and
Isobe et al. (1986).
[http://www.astro.psu.edu/statcodes/asurv]
- asWedit
- A comprehensive and easy-to-use HTML3, HTML2, and text editor for the
X Window System with
Motif.
It offers a standard text editing mode and two context-sensitive,
validating modes for authoring HTML documents. It has a
graphical user interface with multiple editing windows and
all standard editing features besides the HTML-specific
features.
There are
currently (5/96) versions for nine languages.
The HTML features include the context-sensitive modes in which only
tags valid in the current context are available, full support
for all HTML 2 and 3 tags, support in HTML 2x and 3x extended
modes for such elements a new tables, client-side image maps,
etc.), support for Netscape extensions such as frames,
creation of correct documents from editor input,
support for different editing styles (e.g. assistive tagging,
cut and paste, by hand and parse), dialogs for selecting relative
and absolute URLs, customizable colors for different HTML tags,
text to table or list converters, a table of contents generator,
ready to use examples, spell checking, user definable key bindings,
comprehensive and context-sensitive hypertext help, a preview mode,
and much more.
AsWedit is available free for use in non-profit institutions.
It is available in binary form for IBM RS/6000, DEC Alpha,
HP9000 700/800, SGI, SunOS, Ultrix, SCO, and Linux platforms.
[ftp://src.doc.ic.ac.uk/packages/www/asWedit/index.html]
[ftp://ftp.ruf.uni-freiburg.de/pub/info/asWedit/
]
- Atari800
- An emulator for the Atari 800,
800XL, 130XE, and 5200 game machines.
The features include:
versions for X11, SVGALIB, Amiga, MS-DOS, curses, and dumb terminals;
a GUI interface for Xview,
Motif, and Amiga versions;
a menu system for all versions that support bitmapped graphics;
handling standard 8K and 16K as well as OSS super cartridges;
display list interrupts;
a PIL mode;
GTIA graphics support for all display modules;
horizontal and vertical fine scrolling;
replacement of cassette device with host device to give access to host
file system;
reading XFD and ATR disk image formats;
sound emulation; and
printer support.
[http://www.signus.demon.co.uk/david/atari/atari.html]
- ATF Tools
- The Automated Telescope Facility Tools
is a collection of tools written for use at the Automated Telescope
Facility at the University of Iowa. The programs are mostly
utilities to analyze and manipulate
FITS image files although there are also
programs for differential and absolute photometry and
astrometry.
The programs in ATF Tools include:
- FITShdr, which displays and edits headers in FITS format files;
- WCS, which displays, adds, and edits World Coordinate System
(WCS) keywords in FITS headers;
- Crop, which aligns and crops FITS images using either a
specified bounding box or a common area;
- F Compress, which compresses FITS files using the H algorithm;
- F Decompress, which decompresses files compressed using
F Compress;
- Photom, a differential photometry package;
- Fphotom, for field aperture photometry of images;
- Photcal, which solves for air mass and instrumental constants
using standard field images;
- Predict, which predicts times of minima using an input
ephemeris table;
- SNSearch, which searches for new stars by comparing
archive images; and
- Vsmon, which performs automated variable star monitoring using
differential photometry.
The source code for ATF Tools is available. It is written
in ANSI C and can be compiled and used on a wide variety of
UNIX and non-UNIX platforms.
It is also available in binary format for Linux platforms.
The programs are documented in man pages.
[http://www-astro.physics.uiowa.edu/]
- Athena Widget Set
- See Xaw.
- ATM
- This is a project to provide
Asynchronous Transfer
Mode support for Linux.
An experimental release supports
raw ATM connections and basic IP over ATM along with
preliminary signaling. See the site for explanations of the
alphabet soup in the previous sentence.
See McDysan and Sophn (1994),
Schatt (1996), and
Taylor (1995).
[http://lrcwww.epfl.ch/linux-atm/info.html]
- AtmosThermoPack
- A package of Fortran subroutines for the analysis and
calculation of various thermodynamic quantities in the
atmosphere.
There are routines to calculate the:
thermal conductivity of air,
specific heat of ice,
specific heat of moist air at constant pressure,
specific heat of water,
temperature derivative of saturation vapor pressure
over water,
temperature derivation of saturation vapor pressure
over ice,
entropy referred to the triple point and 1013 mb,
enthalpy,
dry static energy,
moist static energy,
water vapor pressure,
saturation vapor pressure over water,
saturation vapor pressure over ice,
latent heats of water from empirical fits,
altitudes or pressure from the dry hydrostatic equation,
saturation mixing ratio,
atmospheric density of moist air,
relative humidity,
mixing ratio,
specific humidity,
dew-point temperature,
frost-point temperature,
lifting condensation level temperature,
moist adiabatic lapse rate,
virtual temperature,
virtual temperature including water content loading,
dry potential temperature,
moist potential temperature,
virtual potential temperature,
equivalent potential temperature,
temperature at a desired pressure level,
temperature from equivalent potential temperature,
temperature from virtual potential temperature with liquid condensed phase,
dynamic viscosity of air, and
kinematic viscosity of air.
A source code distribution of AtmosThermPack is available.
It is written in Fortran 77 and is documented via comment
statements contained within the source code file.
[ftp://climate.gsfc.nasa.gov/pub/wiscombe/Atmos_Thermody/]
- Acorn Atom Emulator
- An emulator for the Acorn Atom computer.
[http://wwwis.cs.utwente.nl:8080/~faase/Ha/Atom/]
- a2ps
- An ASCII to PostScript converter with
extended pretty-printing capabilities.
The default format used is two pages on each physical page,
borders surrounding pages, headers with useful information (e.g.
page number, printing date, file name), line numbering,
pretty-printing, symbol substitution, etc.
Style sheets are included for a wide range of languages
which guide how the text is to be printed and are extensible by
the user, with automatic style selection via another modifiable file.
It is possible to delegate the processing of some files to other
filters or programs via a configurable script, e.g. using
TeX or groff to
process a file containing source code in either markup language.
Currently (10/97) available are DVI, compression, HTML,
PostScript, Roff, and Texinfo filters.
A source code distribution of a2ps is available. It is written
in C and can be compiled and used on most UNIX flavors.
A user's manual is included in the distribution.
Some ancillary packages are also available including:
a2print, a graphical interface to a2ps;
nh2ps, a Korean version of a2ps;
a2ps-greek, a Greek version of a2ps; and
Okonkify, a program that extends fonts designed for Latin 1 so
they support other Latin encodings.
[http://www-inf.enst.fr/~demaille/a2ps/]
- AUC TeX
- See under Emacs.
- AudioFile
- A network-transparent audio server and client library.
AudioFile is also device-independent and allows multiple audio
applications to be run simultaneously, sharing access to the actual
audio hardware.
The network transparency means that application programs can run on
machines scattered through a network, and the device independence means
that applications don't have to be rewritten to work with new audio
hardware. An analogy would be that it does for sound what X11 does
for text and graphics.
The AF distribution includes device drivers for several devices,
server code for a number of platforms, a programming API and
library, out of the box core applications, and several contributed
applications.
A source code distribution is available at the original DEC site,
although porting it to Linux may be more than trivial.
A binary distribution for Linux Intel can be found at the MIT
site.
Documentation is scattered throughout the distribution.
[http://www.tns.lcs.mit.edu/vs/audiofile.html]
[http://gatekeeper.dec.com/pub/DEC/AF/
]
- AUIS
- The Andrew User
Interface
System, sometimes called just ``Andrew,''
is a compound document environment offering a word processor,
a mail/bulletin board reader/writer, a drawing editor, a spreadsheet,
a font editor, an application builder, and other facilities. It
is available separately or wholly in the files of the form
auis*-+.tgz where the + indicates one of the following: wp - the
basic word processor; src - the source code
for developers; mail - the MIME-compatible mail interface; or
full - all of the above.
The basic application in the word processing package is the
ez editor which can be used to edit text and graphics and can
also serve as a word processor.
It loads a document, displays it in a window and automatically
displays to appropriate editing commands, i.e. a text document
will cause it to show text-editing commands and a picture
picture-editing commands. It is not quite WYSIWYG but rather
shows the text and pictures in a form that varies with screen
size and with slightly different fonts. Additional features
include automatic checkpointing, multiple windows, document
output in PostScript
and RTF formats, many choices for
text formatting styles as well as the capability to easily
create your own, an extensive set of document templates, and
tools to facilitate the editing of source code in many
languages.
The basic mail reading tool is called messages, which can serve as a
conventional mail user agent as well as support reading and posting
to bulletin boards and delivery between cells in the Andrew File
System (AFS). This offers a GUI for sending, reading, editing and
printing mail. Messages is compatible with present ASCII mail
delivery systems and also fully supports MIME extensions.
Other tools in AUIS include bush, a graphical interface to the
file system; chart, which allows you to create simple graphs some
numerical data; figure, a drawing editor; raster, an editor for
digitized pictures; typescript, an alternative to xterm; prefed,
a specialized preferences editor; and much more.
The documentation
is written in an internal format and over 800 KB worth can be
perused using the Andrew tools. You can also obtain a hard copy
from the developers for around $30. For introductory purposes you
can obtain PostScript
versions of four articles that ran in the
Linux Journal from Aug. to Nov. 1994.
The source code is available as well as
binary distributions for Linux, HP, Sun, Dec Ultrix and IBM RS6000
platforms. The compressed binary distributions are around 10 Mb
compressed and 40 Mb uncompressed.
[http://www.cs.cmu.edu/afs/cs.cmu.edu/project/atk-ftp/web/andrew-home.html]
- A'UM
- An implementation of the A'UM concurrent object-oriented programming
language, a descendant of the KL1 language (implemented in
KLIC).
The language features include objects which all run concurrently,
stream communication in which objects intercommunicate via directed
paths (i.e. streams) with operations on streams providing all the
necessary communications,
fine grain parallelism, and
intricate synchronization.
The system provides a environment for concurrent execution of
A'UM programs on UNIX platforms, a foreign language
interface which enables programs written in other languages to
be combined with A'UM programs, and a debugger
furnished with breakpoints, stepping and tracing execution, and
several other functions.
A source code distribution of A'UM is available. It is written
in C and can be compiled on most generic UNIX platforms (with
GCC 1.37 or greater recommended).
Various manuals are available although all are written in
Japanese.
[http://www.icot.or.jp/AITEC/IFS/IFS-abst/005.html]
- AutoClass C
- AutoClass is an unsupervised Bayesian classification
system which seeks a maximum
posterior probability classification.
The key features include: automatic determination of the
number of classes, use of mixed discrete and real valued data,
capability of handling missing values, processing time roughly
linear in the amount of data, cases which have probabilistic
class membership, correlation between attributes in a class,
generation of reports describing the classes found, and
prediction of test case class memberships from a training
classification.
A database of attribute vectors (i.e. cases) is input along with
a class model. AutoClass finds the set of classes which is maximally
probable with respect to the data and model and outputs a set of
class descriptions along with partial membership of the cases in
the classes.
AutoClass C is an implementation of the AutoClass algorithm written
in ANSI C which is about 10 to 20 times faster than the original
Lisp implementations. It has been ported to and test on
Sun Solaris and SunOS, SGI IRIX, Intel Linux, and HP-UX
platforms.
The algorithm and package are documented in several manuals
and technical reports available in PostScript format.
See also Snob.
[http://ic-www.arc.nasa.gov/ic/projects/bayes-group/group/autoclass/autoclass-c-program.html]
- autoconf
- A utility created for the
GNU project to create
configuration scripts for software packages. A file containing variables
that are aliases
for various testing sequences is created to check for, e.g. the
compiler, libraries, and other software needed to compile a
specific program. This is processed to create a file called
configure which contains the actual commands to perform the
tests which, in turn, creates the makefiles needed to compile
the program. A large variety of built-in variables for various
test sequences is available and the user can create custom-built
variables and sequences as needs be via a standard procedure.
The source code, written in ANSI C, is available for autoconfig
as it is for all GNU packages, and is readily configured and
compiled via the supplied configure script. Keep in mind that
autoconfig creates the configure files so it isn't needed for
a package that comes with a configure file. Most of the time
autoconfig is used by the author of a software package to create
a configure file that can be used to install the software on the
widest possible range of platforms. As such it is seldom needed
or used by those who only compile and use software built by others.
The documentation for autoconfig is contained within a
Texinfo file whose printed
version runs to 100+ pages.
[http://www.gnu.ai.mit.edu/order/ftp.html]
- Automake
- A Makefile generator for automatically creating files called
Makefile.in from files called Makefile.am, with
the former intended to be used by the
autoconf package. The
latter files consist of a series of make macro
definitions along with the occasional rule, and the former
are compliant with the GNU Makefile standards.
The goal of Automake is to remove the burden of Makefile
maintenance from individual GNU maintainers and put it on
the maintainer of Automake.
A source code distribution of Automake is available. It
is written in and thus requires Perl.
It is documented in a user's guide available in
Texinfo format.
[http://www.gnu.ai.mit.edu/order/ftp.html]
- autoson
- A distributed batch queueing tool for scheduling processes
across a network of UNIX workstations.
It enables the execuation of a stream of processes in a flexible
and convenient way with minimum impact on interactive users.
Autoson can be compiled either in single- or multi-user mode,
with the former providing the capability for a single user to
execute processes on one or more workstations and the latter
allowing several users to use the same queue.
It can execute independent processes on heterogenous/unreliable
processors, but does not provide a parallel environment allowing
processes to communicate. It simply executes processes and waits
for them to finish.
Autoson operates by maintaining a queuefile containing
entries comprising queues of tasks that need to be performed.
Each entry has a unique number and is one of six states, i.e.
PENDING, HOLDING, WAITING, CURRENT, LOST, or SICK.
Each entry has a large number of possible attributes controlling
such things as time limits, priorities, niceness, logging,
timestamping, etc.
Six commands are implemented as links to the main executable
autoson. These are:
auadd, which adds new entries to the queue;
aumod, which modifies existing entries;
auzap, which kills the processes running entries;
aurun, which executies entries;
aulook, which examines a queue;
and aulock, which locks the queuefile and possibly edits it.
Autoson can be seen as a simpler and easier-to-use version of
Generic-NQS or
DQS.
A source code distribution of autoson is available.
It is written in C and supports several UNIX flavors including
Sun Solaris, SGI IRIX, DEC OSF1 and Ultrix, HP-UX, and Linux.
It is documented in a 38 page user's manual available in
PostScript format.
[http://cs.anu.edu.au/people/bdm/autoson/]
- awk/AWK
- A pattern scanning and processing language originally
designed for text processing
applications, especially those in which information is structured
in records and fields.
It searches one or more files for a specified pattern and then
performs specified actions each time it finds a match.
The language was defined in Aho et al. (1988).
The GNU version of this called
Gawk is a superset of awk.
The sample code from the book can be found in the
awkbookcode subdirectory at the second URL given below.
A source code distribution of the ``one true awk'' is available
at Brian Kernighan's web site (the first URL below).
It is written in ANSI C and can be compiled on most UNIX
flavors. It compiled right out of the box on my Linux
platform.
See Dougherty (1992) and
Dougherty and Robbins (1997).
[http://cm.bell-labs.com/who/bwk/
[http://www.netlib.org/research/
]
- awk2c
- An Awk-to-C translator based on
Gawk 2.15.6.
This converts Awk source code to the equivalent C source
code. This is linked with a static library to create
a standalone executable for the original Awk program.
[http://sunsite.unc.edu/pub/Linux/utils/text/]
- AwkTools
- A Java regular expression package that
implements the fast DFA-based pattern matching algorithms
used in awk.
A source code implementation of NetComponents is freely available
under the terms of a non-exclusive, non-transferable limited
license whose details are available at the site.
This requires JDK 1.1 or higher and
the related OROMatcher package.
The API is documented in HTML format.
[http://www.oroinc.com/]
- aXe
- A text editor which is designed to be an
improvement over the xedit editor.
It is built around the Athena
(Xaw) text widget and features:
multiple windows and buffers,
a default menu interface,
configurable menus,
an optional configurable button interface,
a minibuffer for expert use and access to internal filters,
provision for defining a keyboard macro,
geometry specification and resizing in terms of characters,
file selection via a browser,
knowledge of line numbers,
parenthesis matching,
searching via regular expressions,
restricted or unlimited undo,
the ability to change fonts,
easy entry of control codes,
an xterm-like keymap feature,
easy runtime setting of selected preferences or resources,
brief and comprehensive online hypertext help,
a server mode with cooperating client programs,
an optional extension language using Tcl,
optional 3-D Xaw widget compatibility, and
a collection of reusable widgets which embody the functionality
of aXe.
A source code distribution of aXe is available as well as
a binary for Linux Intel. It is written
in C and can be compiled and used on
many UNIX platforms.
Use of the full feature set also requires an installation of
Tcl/Tk.
Documentation includes a manual in
PostScript format as well as
a man page.
[http://sunsite.unc.edu/pub/Linux/apps/editors/X/]
- AXIS
- A 3-D network accessible rendering engine/browser which can
be used to build shared virtual environments.
The network support allows you to open a TCP/IP
port and then telnet to it to control the rendering.
The AXIS interpreter is multi-threaded and allows objects
in the 3-D environment to have private namespaces.
A language called PolyScript is used to create the
environments.
It was created and optimized specifically for creating and modifying
3-D objects and for binding events and messages to user interactions
with those objects.
It somewhat resembles PostScript in
terms of syntax.
Binary distributions of AXIS are available for
Linux Intel, Windows 95/NT, SGI IRIX, and Sun Solaris and SunOS
platforms.
It uses either OpenGL or
Mesa to perform the 3-D rendering.
The documentation is a bit sparse although there are the
beginnings of a PolyScript manual included in the distribution.
[http://www.ikm.com/ckqXZaac/]
- Aztec
- An iterative library which simplifies the process of solving
sparse linear systems of equations .
It is intended as a tool for users who want to avoid cumbersome
parallel programming details when solving large sparse linear
systems on parallel computing systems.
The Aztec package also includes a set of data transformation
tools for the easy creation of distributed sparse unstructured
matrices for parallel solution.
Aztec includes a number of Krylov iterative methods such as
CG, GMRES, and BiCGSTAB. These are used in conjunction with various
preconditioners such as polynomial preconditioners or domain
decomposition using LU or incomplete LU factorizations within
subdomains. The sparse matrix can be general although the
package has been specifically designed for matrices arising
from the approximation of PDEs, with the preconditioners,
iterative methods, and parallelization techniques oriented
towards systems arising from PDE applications.
A source code distribution of Aztec is available.
It is written in ANSI C and can be compiled and used on a variety
of platforms. It has been used on several parallel machines
including nCUBE 2, IBM SP2, Intel Paragon, and MPI platforms.
It can also be used on standard serial and vector platforms.
[http://www.cs.sandia.gov/CRF/aztec1.html]
