LSST Data Management: Entering the Era of Petascale Optical Astronomy

Abstract

The Large Synoptic Survey Telescope (LSST; Ivezic et al.2008, http://lsst.org) is a planned, large-aperture, wide-field, ground-based telescope that will survey half the sky every few nights in six optical bands from 320 to 1050 nm. It will explore a wide range of astrophysical questions, ranging from discovering killer asteroids, to examining the nature of dark energy. LSST will produce on average 15 terabytes of data per night, yielding an (uncompressed) data set of 200 petabytes at the end of its 10-year mission. Dedicated HPC facilities (with a total of 320 TFLOPS at start, scaling up to 1.7 PFLOPS by the end) will process the image data in near real time, with full-dataset reprocessing on annual scale. The nature, quality, and volume of LSST data will be unprecedented, so the data system design requires petascale storage, terascale computing, and gigascale communications.