THE AIPS FAQ


Archive-name: astronomy/aips-faq
Posting-frequency: monthly
Last-Modified: 1997/05/29
Version: 2.5
Modified links: 2007/10/29, information is somewhat out of date


INTRODUCTION

The main source for this FAQ (list of Frequently Asked Questions) is AIPS memo 87, "The NRAO AIPS Project --- A Summary" by Alan H. Bridle and Eric W. Greisen. What you are now reading expands in relatively minor ways the contents of that memo. A plain text version of this document is also available via anonymous ftp to ftp.aoc.nrao.edu/pub/software/aips.

This FAQ was posted to alt.sci.astro.aips and other news groups. The bananas exploder is mirrored still to/from this news group but we no longer pay much attention to it.


TABLE OF CONTENTS

  1. WHAT IS AIPS?
  2. WHAT IS IT USED FOR?
  3. WHAT DOES IT RUN ON? WHAT PERHIPHERALS DOES/CAN IT USE?
  4. WHO USES IT?
  5. HOW CAN I FIND OUT MORE?
  6. HOW DO I KNOW IT WORKS, OR HOW FAST OR WELL?
  7. WHAT OF THE FUTURE?
  8. HOW CAN I GET IT - WHO CAN I CALL?
  9. IS THERE A FAQ FOR INSTALLERS and MANAGERS?
  10. IS THERE A FAQ FOR USERS?


1. WHAT IS AIPS?

For a quick look, check this out.

The NRAO Astronomical Image Processing System (AIPS) is a software package for interactive (and, optionally, batch) calibration and editing of radio interferometric data and for the calibration, construction, display and analysis of astronomical images made from those data using Fourier synthesis methods. Design and development of the package began in Charlottesville, Virginia in 1978. It presently consists of over 4300 files containing 1.46 million lines of text. These comprise over 400,000 lines of documentation and on-line help in over 1300 files, and almost a million lines of text in over 2300 Fortran and C source files. It contains over 350 distinct applications "tasks," representing well over 60 person-years of effort since 1978.

At its peak, The AIPS group in Charlottesville and Socorro had five full-time scientist/programmers, and a few other computing and scientific staff with partial responsibility to the AIPS effort. The size of the group has shrunk considerably since then. The group is responsible for the code design and maintenance, for documentation aimed at users and programmers, and for making the code available to non-NRAO sites. Since its release under the Free Software Foundation's General Public License in mid-1995, its availability via the internet has been made considerably easier, and for that one release (15JUL95), about 150 sites downloaded the software, and conservative estimates of the number of machines running this one version of AIPS alone indicate about 450. NRAO currently offers AIPS installation kits (ready-to-run binaries) for most of the currently available UNIX systems, with updates available semi-annually. The number of sites running some version of AIPS is probably in excess of 250, possibly more than 300.

In 1983, when AIPS was selected as the primary data reduction package for the Very Long Baseline Array (VLBA), the scope of the AIPS effort was expanded to embrace all stages of radio interferometric calibration, both continuum and spectral line. The AIPS package contains a full suite of calibration and editing functions for both VLA and VLBI data, including interactive and batch methods for editing visibility data. In 1996/7, considerable effort was expended to ensure that AIPS would be capable of handling the data from Orbiting VLBI satellites such as VSOP. For VLBI, it reads data in MkII, MkIII and VLBA formats, performs global fringe-fitting by two alternative methods, offers special phase-referencing and polarization calibration, and performs geometric corrections, in addition to the standard calibrations done for connected-element interferometers. The calibration methods for both domains encourage the use of realistic models for the calibration sources and iterated models using self-calibration for the program sources.

2. WHAT IS IT USED FOR?

AIPS has been the principal tool for display and analysis of both two- and three-dimensional radio images (i.e., continuum "maps" and spectral-line "cubes") from the NRAO's Very Large Array (VLA) since early in 1981. It has also provided the main route for self-calibration and imaging of VLA continuum and spectral-line data. It contains facilities for display and editing of data in the aperture, or u-v, plane; for image construction by Fourier inversion; for deconvolution of the point source response by Clean and by maximum entropy methods; for image combination, filtering, and parameter estimation; and for a wide variety of image and graphical displays. It records all user-generated operations and parameters that affect the quality of the derived images, as "history" files that are appended to the data sets and can be exported with them from AIPS in the IAU-standard FITS (Flexible Image Transport System; see newsgroup sci.astro.fits format. AIPS implements a simple command language which is used to run "tasks" (i.e., separate programs) and to interact with text, graphics and image displays. A batch mode is also available. The package contains over 7 Megabytes of "help" text that provides on-line documentation for users. There is also a suite of printed manuals for users and for programmers wishing to code their own applications "tasks" within AIPS.

3. WHAT DOES IT RUN ON? WHAT PERHIPHERALS DOES/CAN IT USE?

An important aspect of AIPS is its portability. It has been designed to run, with minimal modifications, in a wide variety of computing environments. This has been accomplished by the use of generic FORTRAN wherever possible and by the isolation of system-dependent code into well-defined groups of routines. AIPS tries to present as nearly the same interface to the user as possible when implemented in different computer architectures and under different operating systems.

The NRAO has sought this level of hardware and operating system independence in AIPS for two main reasons. The first is to ensure a growth path by allowing AIPS to exploit computer manufacturers' advances in hardware and in compiler technology relatively quickly, without major recoding. (AIPS was developed in ModComp and Vax/VMS environments with Floating Point Systems array processors, but was migrated to vector pipeline machines in 1985. Its portability allowed it to take prompt advantage of the new generation of vector and vector/parallel optimizing compilers offered in 1986 by manufacturers such as Convex and Alliant. It was extended in simple ways in 1992 to take full advantage of the current, highly-networked workstation environment.) The second is to service the needs of NRAO users in their home institutes, where available hardware and operating systems may differ substantially from NRAO's. By doing this, the NRAO supports data reduction at its users' own locations, where they can work without the deadlines and other constraints implicit in a brief visit to an NRAO telescope site.

The exportability of AIPS is now well exploited in the astronomical community; the package is known to have been installed at some time on a large number of different computers, and is currently in active use for astronomical research at somewhere around 250 sites worldwide (Reference ** below indicated 140, but qualitatively the number is now guesstimated to be considerably higher). AIPS has been run on Cray and Fujitsu supercomputers, on Convex and Alliant "mini-supercomputers," on the full variety of Vaxen and MicroVaxen, and on a wide range of UNIX workstations including Apollo, Data General, Hewlett Packard, IBM, MassComp, Nord, Silicon Graphics, Stellar and SUN products. It is available for use on Intel-based personal computers under the freely available Linux operating system (Linux, like AIPS, is covered by the GNU General Public License). In late 1990**, the total computer power used for AIPS was the equivalent of about 6.5 Cray X-MP processors running full-time. It is now enormously more than that since ordinary PCs are equivalent to the X-MP.

AIPS is made available either as source code -- where a complete compile/link cycle is needed -- or source plus binaries for a variety of Unix systems. Either form can be obtained via anonymous ftp or, using the installation procedure install.pl, via rsync and cvs. Binary versions are available for SUN Solaris, Linux in 32-bit and 64-bit versions, and Lac OS/X in PPC and Intel chip versions.

Similarly, but somewhat of mere historic reference now, a wide range of digital TV devices and printer/plotters has been supported through AIPS's "virtual device interfaces". Support for such peripherals is contained in well-isolated subroutines coded and distributed by the AIPS group or by AIPS users elsewhere. Interactive image display in now provided directly on workstations using an AIPS television emulator and X-Windows. Hardware TV devices are now practically extinct, but those used at AIPS sites in the past have included IIS Model 70 and 75, IVAS, AED, Apollo, Aydin, Comtal, DeAnza, Graphica, Graphics Strategies, Grinnell, Image Analytics, Jupiter, Lexidata, Ramtek, RCI Trapix, Sigma ARGS, Vaxstation/GPX and Vicom. With Printer/plotters, in the age prior to PostScript becoming almost universally accepted as the language of choice for these devices, AIPS support included Versatec, QMS/Talaris, Apple, Benson, CalComp, Canon, Digital Equipment, Facom, Hewlett-Packard, Imagen, C.Itoh, Printek, Printronix and Zeta products. Generic and color encapsulated PostScript is now produced by AIPS for a wide variety of printers and film recorders. The standard interactive graphics interface in AIPS is the Tektronix 4012, now normally emulated on workstations using an AIPS program and a terminal emulator such as xterm under X-Windows.

4. WHO USES IT?

The principal users of AIPS are VLA, VLBA, and VLBI Network observers. A survey of AIPS sites carried out in late 1990** showed that 61% of all AIPS data processing worldwide was devoted to VLA data reduction. Outside the NRAO, AIPS is extensively used for other astronomical imaging applications, however. 56% of all AIPS processing done outside the U.S. involved data from instruments other than the VLA. The astronomical applications of AIPS that do not involve radio interferometry include the display and analysis of line and continuum data from large single-dish radio surveys, and the processing of image data at infrared, visible, ultraviolet and X-ray wavelengths. About 7% of all AIPS processing involved astronomical data at these shorter wavelengths, with 7% of the computers in the survey using AIPS more for such work than for radio and another 7% of the computers using AIPS exclusively for non-radio work.

Some AIPS use occurs outside observational astronomy, e.g., in visualization of numerical simulations of fluid processes, and in medical imaging. The distinctive features of AIPS that have attracted users from outside the community of radio interferometrists are its ability to handle many relevant coordinate geometries precisely, its emphasis on display and analysis of the data in complementary Fourier domains, the NRAO's support for exporting the package to different computer architectures, and its extensive documentation.

5. HOW CAN I FIND OUT MORE?

As well as producing user- and programmer-oriented manuals for AIPS, the group publishes a newsletter that is sent to 100's of AIPS users outside the NRAO semi-annually, with one issued timed to the annual "release" of new AIPS code. There is also a mechanism whereby users can report software bugs or suggestions to the AIPS programmers and receive written or email responses to them; this has in the past provided a formal route for user feedback to the AIPS programmers and for the programmers to document difficult points directly to individual users. With the 15JAN96 release, the old "gripe" system will now have the ability to automatically submit problem reports directly via email to the AIPS group within NRAO. This is expected to complement the existing informal approach involving support via electronic mail. In addition, a registration mechanism has been instituted that will enable NRAO to keep track of the distribution of AIPS, as serious sites are expected to register in order to obtain user support.

Much of the AIPS documentation is now available to the World-Wide Web so that it may be examined over the Internet (start with the AIPS home page). Also, this information is available via anonymous ftp. The NRAO knows of over 230 AIPS "tasks," or programs, that have been coded within the package outside, and not distributed by, the observatory.

There is a closed, moderated mailing list called "bananas" that serves as a conduit for important announcements pertaining to AIPS, as well as an occasional forum for questions and discussion about the software. You can subscribe yourself to this list on-line.

6. HOW DO I KNOW IT WORKS, OR HOW FAST OR WELL?

The AIPS group has developed two packages of benchmarking and certification tests that process standard data sets through the dozen most critical stages of interferometric data reduction, and compare the results with those obtained on the NRAO's own computers. These "DDT" (Dirty Dozen Tasks) and "Y2K" packages are used to verify the correctness of the results produced by AIPS installations at new user sites or on new types of computer, as well as to obtain comparative timing information for different computer architectures and configurations. They have been used extensively as a benchmarking package to guide computer procurements at the NRAO and elsewhere. Two other packages, "VLAC" and "VLAL", are less widely used to verify the continued correctness of continuum and spectral-line reductions.

The "AIPSMark(93)" and new "AIPSMark(00)" are often used as measures of the performance that a given machine will produce in real data reduction situations. The first measure, and other aspects of the DDT package, are described in AIPS Memo 85. The new measure and the Y2K package are described in AIPS Memo 104. The original baseline for the AIPSMark was a Sparcstation IPX, set by definition at 1.0. In general, high-end Intel Pentium III 800MHz systems may give an AIPSMark of around 23 (per cpu), mid-range Sparc Ultra systems around 9 and Ultra-60's around 20, and Digital Alpha and SGI processors may produce 15-20 per cpu. The record is 59 set by Compaq alpha with native compiler. It is important to note that the benchmark measures total system performance as it is based on total elapsed time. Thus, factors such as disk latency and transfer rates, memory, swapping, and others are perhaps almost as important as raw floating-point performance.

7. WHAT OF THE FUTURE?

In 1992, the NRAO joined a consortium of institutions seeking to replace all of the functionality of AIPS using modern coding techniques and languages. The aips++ project is expected to provide the main software platform supporting radio-astronomical data processing sometime around the turn of the century. Future development of the original ("Classic") AIPS will therefore be somewhat limited, mostly to calibration of VLBI data, general code maintenance with moderate enhancements, and improvements in the user documentation. Since aips++ has taken a long time to replace AIPS, development in AIPS has continued albeit at a slower rate.

aips++ has been renamed CASA which is now (12/2007) in beta release. More details can be found on the CASA web page at: http://casa.nrao.edu/

8. HOW CAN I GET IT - WHO CAN I CALL?

Since the 15JUL95 release of AIPS, the software is Copyright (C) 1995-2008 by Associated Universities, Inc., and is protected by the Free Software Foundation's General Public License (GPL). It is freely available under the terms of this License on our web and anonymous ftp servers. However, there is a registration mechanism, and only those AIPS installations that have registered with NRAO will be eligible to receive any form of support.

Prior to this release, AIPS was proprietary software issued to various people under what is now an obsolete User Agreement. For academic or educational or research oriented users, there was no charge for the agreement, but there was a fee for commercial users. This is no longer the case.

Why was this so? Why did NRAO/AUI try to control distribution? The answer to both of these is twofold. First, it was labelled as proprietary code to prevent third parties from taking the code (for free), slightly changing it, slapping a copyright on it and sueing NRAO to cease and desist from distributing the original AIPS. While this may sound unlikely, this sort of thing has happened to others. The GNU General Public License now protects NRAO and our users from this sort of scenario in a less restrictive way.

Second, it is really important to us to have a clear picture of how many users of AIPS there are out there. Not only does this give us a certain amount of leverage with hardware and software vendors, but it helps to justify allocation of resources (people, computers) specifically for continued support of AIPS. That is why we have retained some flavour of the old system in the new "register for support" scheme. This has already helped to provide a picture of both the extent of AIPS use, and the type of hardware on which it is being installed, to the AIPS group.

The AIPS group encourages all AIPS "customers" who intend to use the software for Astronomical Research (especially those working in Radio Astronomy) to register with NRAO when they receive and install a copy of AIPS. The installation procedure now does this for you automatically if you approve, and will even e-mail in the registration form to us. Eventually we hope to have a forms-based interface for registering on our web pages (when we find that elusive spare time).

Further information on AIPS can be obtained by writing by electronic mail to aipsmail at nrao.edu or by paper mail to the AIPS Group, National Radio Astronomy Observatory, Edgemont Road, Charlottesville, VA 22903-2475, U.S.A. Use this address to request copies of AIPS or ancillary documentation also. Do not use it for UCE, SPAM, junk e-mail or other net.abuse.

9. IS THERE A FAQ FOR INSTALLERS and MANAGERS?

The best source for information on the AIPS installation wizard is the Running the install wizard guide. It is updated whenever the install.pl script is changed. Another source for information about AIPS Installation is the AIPS Installation Summary. This document used to be updated for every release, but has fallen into disuse with the development of the wizard.

There is a separate, but unmaintained, document called the AIPS Unix Porting Reference. However, this has not been updated since the July 1994 release, and significant portions of it are now either largely irrelevant or have been incorporated into the Installation Summary described above. It might be somewhat useful for ports to new Unix or non-Unix systems. In the future, it is hoped to make this available on-line via regular web pages as opposed to a TeX/PostScript document.

A section on the most frequent problems will be added; this has evolved into a AIPS Manager FAQ.

10. IS THERE A FAQ FOR USERS?

The closest thing to such a document is probably the AIPS Cookbook. It may be found online at http://www.aips.nrao.edu/cook.html in the form of a table of contents and individual PostScript files for each chapter.


[**] The 1990 AIPS Site Survey, AIPS Memo No. 70, (Warning! WordPerfect binary file!) Alan Bridle and Joanne Nance, April 1991


[Powered by Apache!] Eric W. Greisen