Web site

Spica is web-based front-end to the TERAPIX Megacam data reduction pipeline. To use it, you must first register with TERAPIX for a Spica account. From the Spica website, you can then select which data you would like to reduce and what operations to whish to apply to them. Once the processing is terminated, an email alert describing how to retrieve the final data product is sent to the owner's account.

Spica account

Because data reduction is a resource-intensive task, you must first request for an account. Each user can choose a reduction group which should reflect his scientific objectives. For example, if you are interested in processing solar system data, you should use the solar system group. Appropriate configuration files are available for each reduction group. If you forget your password, it can be retrieved automatically.

Special accounts

The "Administrator" group has special privileges and is allowed to build configuration files for all group members. Each group has a separate administrator account and these accounts have higher execution priority than the standard user.

Web interface

Once you login, you have to define a process name which should reflect your reduction task. Default configuration files for each task are available but you can build your own. These files describe which option the different application will use to reduce the data. Also predefined "Rules" files are available. These files describe how applications should be run. For example, you can define a bigger set of data to get a higher level of astrometric accuracy.

Data selection

Task selection

The main goal of Spica is to run data reduction applications. Five are available but more can be added in future.

• SExtractor : source extractor from FITS image
• Astrometrix : compute astrometric solution from a defined catalog
• Photometrix : set relative photometry in a set of images
• Scamp : Astrometry/Photometry
• SWarp : Resampling and coadding of FITS images

2) Configuration files : Group, user

How does it work?

When you have completed all the web forms, PHP scripts will store all your selection in a Mysql database. When you decide to run one of your processid, the script will build few files in order to warn a perl daemon to wake up and process the data. Progress monitoring can be checked on the web site. Once completed, final data are stored and an email is sent. You are then able to watch and check resulting files on the website also. If you are pleased with the output, you can download them on your own computer, if not, feel free to delete or modify your processid to run it again. Depending on the average load, where the selected data are stored on your cluster and your privilege, a priority number is calculated. Automated processes excecuted by the TERAPIX pipeline will have always a higher priority than those launched from the website so don't expect a very fast computation.

Global flowchart (SNOOPPIX, datatransfer, QualityFITS, pre-processing) (cf another article from 'data reduction')

PHP + Perl + Daemon + quota + priority + Elixir

Because your process can take a lot of our hard disk space, you have a limited disk quota. Once your job is completed, you have 15 days to retrieve the resulting files before they will be deleted from our hard disks. Of course, you'll be warmed before files are deleted! Of course, feel free to send us some big hard disks or some money if you want extra disk storage ;)

Installation

Spica have been developed in order to be easily installed on other computers (Unix only, we are talking about serious OSes only ;) A set of RPM package is available but a few configuration is needed to allow spica to run.

FAQ