|
Документ взят из кэша поисковой машины. Адрес
оригинального документа
: http://www.adass.org/adass/proceedings/adass94/ichikawas.html
Дата изменения: Sat Nov 4 01:46:25 2000 Дата индексирования: Tue Oct 2 03:08:27 2012 Кодировка: Поисковые слова: п п п п п п п п п п п п п п п п п п п п п п п п п п п п п п |
2144 kB PostScript reprint
S. Ichikawa
National Astronomical Observatory of Japan, Mitaka, Tokyo 181,
Japan
S. Yoshida
Kiso Observatory, University of Tokyo, Mitake, Kiso, Nagano 397-01,
Japan
M. Yoshida
Okayama Astrophysical Observatory, NAOJ, Kamogata, Asakuchi,
Okayama 719-02, Japan
T. Horaguchi
National Science Museum, Shinjuku, Tokyo 169, Japan
M. Hamabe
Institute of Astronomy, University of Tokyo, Mitaka, Tokyo 181,
Japan
Japanese optical astronomers have not made much effort to archive the data produced with telescopes in Japan because of their low quality. The construction of Japanese 8m telescope, SUBARU, is now in progress. SUBARU will be in operation by the end of this century and is expected to produce a large amount of high quality data which deserves to be archived. This motivated us to develop a data archive system for SUBARU, although we have little experience in developing or maintaining such a system. To overcome this, we decided to develop an archive system for the CCD data available now at the Kiso Observatory (Kiso) and the Okayama Astrophysical Observatory (OAO). This system, named MOKA (Mitaka Okayama Kiso data Archive system), is designed to be a prototype of the SUBARU data archive system, as well as a practical facility for effective archiving of the data produced at Kiso and OAO.
At present, MOKA deals with raw observational data, not calibrated data. The data are produced from Kiso 105cm Schmidt telescope prime focus camera CCD and OAO 188cm telescope Cassegrain focus spectrograph CCD. MOKA offers data search, data preview (quick look), and FITS header view. MOKA does not handle the raw data directly, and the raw data itself is stored in each observatory on tapes or CD-ROMs. MOKA handles header information, and compressed images for previews extracted from the raw data. The structure of MOKA and the data flow are shown in Figure 1.
Figure: Structure of MOKA and data flow.
Original PostScript figure (18 kB)
Unfortunately, we have a severe bandwidth restriction: 9.6kbps to Kiso and 19.2kbps to OAO. Users of MOKA must access NAOJ Mitaka, or visit Kiso or OAO. We anticipate most users accessing MOKA through Mitaka, although MOKA is installed in Mitaka, Kiso, and OAO.
FITS header data and compressed images are created from the raw data at each observatory, and sent to Mitaka, Kiso, and OAO. Information for a search is then extracted from the header data and input to a database management system (DBMS; in this case Sybase). A GUI has been written for MOKA that creates a SQL file according to search constraints set by the user. At present, possible constraints are instruments (observatories), a range of acquisition dates, types of frames, and a position (ranges of R.A. and Dec.). The SQL file is submitted, and the results of a search are stored in a file. The user can see the list of the frames found in the search, preview these images, and view the header data of the selected frame. MOKA sends requests for raw data to each observatory. The raw data are then sent by tape because of the network restriction.
We use highly compressed images for the previewing, to save both disk space
and a CPU overhead. The compression methods used are binning or
discrete sampling, 8 bit-scaling, and gzip. A 
OAO frame
(about 500kB) is compressed to about 10kB, and a 
Kiso frame
(about 2000kB) is compressed to about 6kB. MOKA displays the compressed
images for preview, and shows either a horizontal or vertical profile
at the requested position.
Figure: Sample screen of MOKA.
Original PostScript figure (2104 kB)
We will soon be able to display many images at the same time in the preview mode, and to set more search constraints (wavelength, etc.). As MOKA is a prototype of the SUBARU archive system, we must address several tasks. For example; the handling of raw and calibrated data directly, dealing with data from other instruments or observatories, connecting with other astronomical databases, and so on. We expect our efforts will lead the better design of the SUBARU archive system.
This work was supported in part under the Scientific Research Fund of the Ministry of Education, Science and Culture (06554001) and in part by NAOJ. The authors thank to Prof S. Nishimura and Prof K. Sadakane for continuous encouragement.
2144 kB PostScript reprint