The Data Analysis System for the SUBARU Telescope

Tadafumi Takata, George Kosugi, Yoshihiko Mizumoto

National Astronomical Observatory of Japan (NAOJ) Osawa 2-21-1, Mitaka, Tokyo 181, JAPAN

Yasuhide Ishihara

Fujitsu Limited, 9-3 Nakase, 1-Chome Mihama-ku Chiba, Chiba 261, JAPAN

Abstract:

1. Overview of SUBARU Data Analysis System

The astronomical data produced by the SUBARU telescope will be very large both in product rate (max 20 MB/s) and in total amount (30 TB/yr). We will drown in a ``flood of data'' unless we develop a data analysis system that can deal efficiently with these data. Thus, we are planning to develop such a system in a distributed environment, which will be served by CORBA (Common Object Request Broker Architecture) on a UNIX system by Fujitsu ltd. The requirements for our system are:

1. To calibrate large quantities of data as quickly as possible.
2. To be able to expand the system easily.
3. To use other existing software (IRAF, SAOimage, etc.).
4. To provide users with a comfortable environment for data analysis, which will allow them to write scientific papers with the least amount of effort.
5. To store calibrated data with information about the history of the calibration in a data archival system.

1. Closely link the system with the database and data archival system.
2. Use the OO method (New Reduction Package (like AIPS++, Glendenning 1994)).
3. Use CORBA
   • Distributed Object Computing
   • Wrapping existing Software (IRAF, SAOimage, etc.)
4. Port IRAF onto Super-Computer VX4, a product of Fujitsu ltd.

  
Figure 1: SUBARU Data Analysis System Structure.
Figure 1: PS 61 Kb

The already existing software such as IRAF and SAOtng will be wrapped as objects and connect with CORBA by way of OMA (Object Management Architecture). Some other applications will be treated as objects in a similar way.

2. Prototyping

2.1. Overview

We tried to port IRAF and SAOtng for prototyping our data analysis system for the SUBARU telescope. We are looking for the best way to deal with many CCD frames that need to be calibrated all at once using an observational log and a GUI interface, and then testing its performance. The aims of our prototype are learning CORBA and C++ programming, finding an interface for IRAF, and studying systems for structure and flow. The environment for prototyping is described below.

1. Machine: Sun S4/10
2. OS: Solaris 2.3 & CORBA 1.1 (Trial version based on CORBA 1.1 by Fujitsu)
3. Applications: IRAF 2.10.4, SAOtng 1.0.1b, 3d View (Original), tcl7.3/tk3.6
4. Language: Sparc C 4.0.1 & Sparc C++ 3.0.1

  
Figure 2: IRAF Wrapping Method.
Figure 2: PS 26 Kb

We used SAOtng as an image browser by wrapping it with XPA (X Public Access) and an original 3d viewer wrapped by means of a pipe as a 3-dimensional plotter. The comparison of our IRAF wrapping method and that of ASSIST by Mandel & Tody (1995) is shown in Table 1.

Table 1: Comparison of IRAF Wrapping Method.

2.2. Merits of our Prototype

1. Can use observational log as the data input list.
2. Can see all operations applied to the data in a log file. (This means the possibility of re-analysis from the middle point of a previous step.)
3. The methods which may be changed can be easily added to the system using window method.

2.3. Problems of our Prototype

1. Too late to perform analysis of some sets of data. (This may be caused by the slowness of initializing the windows written in Tcl/Tk.)
2. There are some IRAF tasks that do not work well---those that have interactive graphic cursor options.
3. The pseudo terminal is a difficult program.
4. CORBA is not developed enough yet, and we will need to wait for future enhancements.

3. Future Plans

• We will wrap IRAF using the ASSIST method.
• The present system needs some refinement to its architecture.
• We will prototype the system with a close linkage to the database system as soon as possible.

References:

Glendenning, B. E. 1994, in Astronomical Data Analysis Software and Systems III, ASP Conf. Ser., Vol. 61, eds. D. R. Crabtree, R. J. Hanisch, & J. Barnes (San Francisco, ASP), p. 413