|
Report of HEASARC Users Committee
January 19, 1995
Summary
The User Group heard an impressive catalog of developments, improvements,
and a solid record of achievement by the HEASARC over the past year.
We find the HEASARC to be a model of efficiency and scientifically
well directed priorities. The unanimous conclusion of the User
Group is that the HEASARC has performed its work remarkably well
on the limited funds available, and with an ever increasing burden
of responsibility.
We find the HEASARC is far more than the sum of its parts. In
addition to several terabytes of archival data (continually
expanding in ingest, and continually increasing in export requests),
it is a unique and invaluable source of assistance (through
software tools and personal assistance/troubleshooting)
without which much research in High Energy Astrophysics would
be impossible. All of this is achieved at only a slight
overhead increase over the cost of maintaining a pure archival
activity.
In the following sections we record our formal recommendations,
comments on the presentations, and summary of the executive
session.
Recommendations & Comments
-
For all datasets being provided to the HEASARC through the ADP
process, proposers should be required by the NRA to adhere
to standard HEASARC data formats.
-
When proposals are accepted which will send data to the HEASARC,
that information should be transmitted to the HEASARC by NASA
Headquarters as early as possible, to allow for resource allocation
at the HEASARC. (We also suggest the possibility of technical
evaluation of proposals for relevant proposals by the HEASARC.)
Sethanne Howard and Guenter Riegler both suggested that
the HUG develop wording to implement the above two points, for
inclusion in the NRA. I will distribute a draft for comment.
-
Use of the NSSDC archive as both the final archive and the
operational archive for XTE is an excellent use of facilities
to maximize the overall NASA efficiency. We encourage the
NSSDC and HEASARC to enlarge areas of joint activity, while
retaining the viewpoints for which each was created (secure
archival storage for the NSSDC, and active science use of
archival material for the HEASARC.)
We recommend to NASA HQ that the project only be allowed to
purchase archive hardware sufficient to store one year of
data for caching purposes.
-
The ASC is critical to future high energy astronomy data
archiving and analysis. We encourage cooperation between
HEASARC and the ASC, and suggest inclusion of ASC personnel
in the HUG and FITS committees.
We specifically recommend the HEASARC coordinate development
of a broader FITS oversight committee (based on the successful
internal one) which reaches out to all major high energy
astrophysics organizations, to ensure the broadest possible
commonality of data and software.
-
FITSIO is an important component in the effort to standardize
data formats and software. The main negative comment has been
a claim that FITSIO is less efficient than other packages.
We recommend that the FITSIO team compile benchmarks, and attempt
to find people with quantitative data about any FITSIO inefficiency.
-
Distributing data on CD-ROMs is viewed as useful scientifically,
and useful as an advertising medium for the HEASARC. Within
the current level of activity we endorse the CD-ROM distribution.
Our suggested priority for future CD-ROMs is imaging data, which
is less efficient to be retrieved by network. The next two
projects are the Einstein MPC data (which are spectra),
and a `Best of Exosat' (which is a mixture of compressed images
and spectra).
-
XSPEC has become an accepted standard for X-ray spectral analysis.
Papers using it are common, but are forced to be improper by
the lack of a published reference. We requested Keith Arnaud
write a paper in a refereed journal, describing XSPEC, which
can be used as an XSPEC citation.
-
The HEASARC has been greatly successful in obtaining, archiving
and reformatting archival datasets to make them publicly useable.
While we recognize that the following datasets are generally
inaccessible for political reasons the HUG calls on the HEASARC,
NASA and our respected colleagues across the world to make
available to the HEASARC the following data.
-
ROSAT All-Sky Survey Data
This is a great observatory with a tremendous wealth of data.
It can only be properly used if it is studied by a much larger
number of scientists. Moreover future X-ray measurements
may be compromised if this archival dataset remains inaccessible.
-
Gamma-ray Burst Data from PVO, Ulysses, Ginga, ISEE-3
The study of Gamma-ray bursts is an extremely exciting
topic. The number of events is small, but these data have been
in hand long enough to reach a non-proprietary status.
Note added: Kevin Hurley has just delivered the Ulysses
data to the HEASARC. Here are Kevin's notes on the data:
For your information, we have just delivered almost 4
years of data to the NSSDC. This consists of high time resolution
(0.25-2.0 s resolution), real-time 25-150 keV count rates, in which almost
all the bursts may be found. The NSSDC negotiated for this data and
actually did not want the higher time resolution (8 or 32 ms) triggered
data. If there is a strong interest in obtaining the triggered data, we
could discuss it. (Please note that, as an ESA/NASA mission, no obligation
was placed on us to deliver anything more than the data we are giving to the
NSSDC.) The only problem we might have is that, like everyone else, we are
being ramped down, so we could not provide a great deal of support.
-
Sigma/Granat & Sigma/ART-P
We are concerned that the storage facilities for these data
in Russia may be inadequate. The HEASARC should investigate
with the Russians and French if the HEASARC can play a role
in saving these data.
-
Steve Drake described the data access and retrieval from
the HEASARC, and commented on the rapid growth of WWW
access. A component of this growth is due to highly
inexperienced or naive users who ask extremely elementary
questions. The HUG encourages the growth of the use of
HEASARC data in educational applications, but suggests
that an education hotseat helpline be developed, either with
money from educational initiatives, or by cross-linking
to an existing educational hotseat.
-
The HEASARC should strive to make accurate, up-to-date
observing logs available as soon as possible for the
active missons (currently ASCA and ROSAT). It would be
desireable to have accurate short term timelines from
the mission schedules, but if mission communication
problems preclude this, then at least a log containing
the processed data should be as current as the processing
effort.
-
When the HEASARC alters FITS format keywords in archives,
or data archive
filenames, it should actively inform the community in advance
of such changes, and consider the community response before
making the change.
-
The HEASARC should add a `What's New' hypertext link to
the HEASARC home page, and update it periodically with
the latest dataset additions, software changes or other
developments. We ask that this link be updated shortly
before the HUG meeting, so that the HUG can test the
new features.
-
The HEASARC should consider implementing a tool similar to
`Presto' in use at the HST. This will track the current
state of an accepted proposal, and provide pointers to
the reduced data products.
Minutes
The following are brief notes from each presentation
-
Nousek- Kirk Borne joins the HUG, replacing Knox Long.
He will keep the HUG in touch with HST archiving activities.
-
White- Big challenge to HEASARC is prospective XTE archive
(3.5 Tb!) XTE will archive directly to NSSDC, without a
project archive.
-
Whitlock- Great progress on data archiving and reformatting.
See earlier recommendations to ease this task.
-
Pence- New checksum keyword added to FITS archive data to
enable data verification.
-
Arnaud- New release of XSPEC (v9.0) due out soon.
-
Drake- Large volume of data requests, primarilly ftp,
with WWW use increasing. Gamma-ray burst data summary.
-
Corcoran- ROSAT archive running well.
-
White- ASCA archive has a new excellent WWW access tool.
-
Calvo- Development of meta-database of pointers to HEASARC
data in early stages.
-
Behnke- Described new upgrades to NSSDC NDADS system, including
DLT jukebox and FDDI network upgrade.
-
Riegler- MO& faces a factor of two reduction in 98-99.
In two years another senior review will compare not only
the mission reviewed recently, but also the grants programs,
LTSA, and the HEASARC. 25% of MO& is spent on mission
planning and operations; he challenges us to improve this by
increased automation and efficiency.
Highlights of the HEASARC
As user groups often focus on the problems encountered over
the past year, we wish to balance the impression, by recognizing
the successes. We list those parts of the HEASARC which we
especially wish to recognize for excellent achievement.
-
The entire WWW-interface to the HEASARC is impressive,
and provides an excellent example to the community.
-
The SkyView system and its WWW access tool.
-
The improvement in on-line accessibility to HEASARC holdings.
-
The great progress on acquiring and restoring datasets.
-
The ASCA archive opened on-time and with no important problems.
-
The HEASARC staff have been highly responsive to user questions.
-
The FITS support work for the community.
-
The international scope and cooperation achieved is a sign
of the respect and quality work done by the HEASARC.
Business for the Next Meeting
The next meeting of the HUG will be planned for early November.
The users prefer a Friday to have lower cost air fare.
Two members will rotate off the HUG, Martin Elvis and Kevin
Hurley. Replacements will be appointed soon.
We request a presentation on the data compression standards
in use at the HEASARC at the next meeting. We also request
a brief report from the NSSDC on the data protection and
multiple copies policies of the HEASARC data held by the NSSDC.
HEASARC user group members can explode messages to the HUG
by addressing them to `hug@astro.psu.edu'.
HEASARC Home |
Observatories |
Archive |
Calibration |
Software |
Tools |
Students/Teachers/Public
Last modified: Wednesday, 24-Oct-2007 09:33:37 EDT
|