An Efficient Retrieval Method for Astronomical Catalog Time Series Data

Astronomical catalog time series data refer to the data collected at different time, which can provide a comprehensive understanding of the celestial objects’ attributes and expose various astronomical phenomena. Its retrieval is indispensable to astronomy research. However, the existing time series data retrieval methods involve lots of manual work and extremely time-consuming. The complexity will also be augmented by the exponentially growth of observation data. In this paper, we propose an automatic and efficient retrieval method for astronomical catalog time series data. With the goal of identifying the same celestial objects time series data automatically, a cross-match scheme is designed, which labeled a unique MatchID for each record matched with the datum catalog. To accelerate the matching process, an in-memory index structure based on Redis is specially designed, which enables matching speed 1.67 times faster than that of MySQL in massive amounts of data. Moreover, Catalog-Mongo—an improved database of MongoDB—is presented, in which a Data Blocking Algorithm is proposed to improve the data partitioning of MongoDB and accelerate query performance. The experimental results show that the query speed is about 2 times faster than MongoDB and 7.6 to 8.7 times than MySQL.