Welcome to

Atlasmaker

What is an Atlas?

An astronomical image contains metadata that places each pixel accurately on the sky through a map projection. In general, every image from a survey has a different projection, but powerful software is now available to reproject these images to a uniform system of projections -- like the pages of an atlas. When diverse image surveys at different wavelengths are all brought to the same atlas, new scientific possibilities arise: finding fainter sources, chracterizing non-pointlike phenomena such as the interstellar medium, and so on. Images can be stacked, like the multi-channel images common in Earth Science, and the data mining algorithms developed there brought to bear on the federated atlas.

A federated atlas of the sky can also be used to create a magnificent resource for education and outreach. Galaxies can be shown by the million instead of as isolated pictures, and students can control their own flight through the sky at multiple scales and multiple wavelengths. This large, carefully controlled, coherent data product can be the foundation for many lessons in math and IT as well as astronomy. The prototype for this is Virtual Sky

Obtaining the code

You can get the Atlasmaker distribution here, it runs with Montage 1.5 and later

Download atlasmaker-1.3.tar
Then follow the instructions on the
Instructions Page.

What is Atlasmaker?

Atlasmaker is a prototype grid-based workflow manager for building atlases of astronomical images. It is designed for high throughput on distributed supercomputing facilities such as Teragrid. It is built from components that include Montage, the NPACI SRB, the NVO Image Access protocol, and the Hyperatlas standard.

This package uses and relies on the compute modules in the core Montage code. It provides an executive to run the whole mosaicking machinery, including background estimation and subtraction. It also prints timings of serial and parallel computing, as well as data fetching times.

Atlasmaker can run on a Unix workstation, or on a parallel machine. Parallelism is through MPI (Message Passing Interface), and assumes that each processor can see the same file space. Atlasmaker can build scripts suitable for being queued in a PBS batch system. The next version of Atlasmaker will include a Condor-G job submission to allow dynamic selection of where the job is run on the Grid.

The workflow environment provides four basic scripts, and two more that use these to generate single mosaics (colorPicture.py) or tiles of an atlas (makeTiles.py). The link from each of these four provides the detailed argument lists.

makeTiles: This is a master script that uses some or all of the components below (makeChart, getData, Run, utilities) to build a set of "tiles" that fit on a page of an atlas.
colorPicture: This is a master script that uses some or all of the components below (makeChart, getData, Run, utilities) to build three mosaics, then combine them into a color image. The script can of course be changed to create only one mosaic.

These two master scripts utilize the functionality below:

makeChart: To make a FITS header (also called a Chart) that defines the output image. This module can build charts from standard atlases according to the Atlasmaker specification, which are therefore guaranteed to federate effectively.
getData: To fetch the relevant data for a given Chart from any NVO image service. Such services now exist for the Sloan Digital Sky Survey, DPOSS, the Digitized Sky Survey, and many others. When 2MASS is released, it will also have NVO interfaces that deliver data through SRB (Storage Resource Broker).
Run: To run Atlasmaker either as a serial or MPI-parallel job, and generate a mosaic. The compute-intensive part, the projection, is parallelized, and the other 10% runs serially.
Utilities: A set of utilities does simple post-processing of the FITS files that Atlasmaker generates. mReduceFloat halves the size by changing 8-byte double pixels to 4-byte float pixels. mReduce makes FITS files with one byte per pixel, while retaining maximum information. mCombine takes three FITS files and builds a color image from them.

Each of these functions is an executable program with arguments, and can therefore be used independently of the others. Atlasmaker also prints timings for both serial and parallel parts of the code.

Documentation

The Instructions Page explains basic principles, and different ways to use the components listed above. It explains a script called colorPicture.py, which makes a set of three mosaic images, then compisits them to a Jpeg. The other high-level script is makeTiles.py, which builds atlas pixels for a standard atlas. Each of these can be modified by users as a basis for their own applications.

The Script Definition page defines the three Python scripts and the utilities mentioned above.

The System Requirements and Fixes document lists the system requirements, and also fixes for soem dommon problems.

Sample Images

Here are a couple of images from the 2MASS survey rendered with Atlasmaker. In each case, the central galaxy is NGC 5371; there is a version with and without background correction.
Note: These image products are not endoresed by the 2MASS project, and are intended as a proof of concept of the Montage algorithm rather than a science product.

The URL of this page is http://www.cacr.caltech.edu/projects/nvo/atlasmaker/.

Atlasmaker technical contributors include the Montage team and: Leesa Brieger (UC San Diego), Mike Feldmann (Caltech CACR), George Kremenek (UC San Diego).