IUCAA Brochure 2024

High Performance Computing For simulations, researchers in IUCAA are currently actively pursuing several topics of cutting edge research To get scientific knowledge out of Astronomical observations we need large computing resources to analyse the raw data and realistically simulate the observations to extract the physical insights. The most common High Performance Computing (HPC) usage in Astronomy and Astrophysics is in i) Simulations of astrophysical phenomenon requiring large scale distributed computing with fast interconnect speeds between nodes and ii) Analysis of Terabyte / Petabyte scale datawhich aremassively parallel and often require fast I/O without the need of fast exchange between nodes. There is also a progressive increase in the application of Machine Learning, suitable for GPU based hardware. For all these domains, the academic members and Associates of IUCAA are already actively involved in cutting edge researchwhich uses state of the art computing codes and are leadingmajor projects and collaborations of international stature. involving HPC in fields such as Galaxy evolution and structure formation, Astrophysical Turbulence, Accretion discs, Numerical relativity, Solar convection, Radiation transport etc. IUCAA also hosts data from various prominent Astronomical observatories, namely, LIGO, Virgo and KAGRA, the MeerKAT Absorption Line Survey (MALS), etc. It also hosts the science support cell for AstroSat. There are dedicated computing clusters for in-house analysis of these datasets. The Sarathi cluster, primarily dedicated to gravitational wave data analysis, offers more than 500TeraFlops of peak theoretical performance. IUCAA also offers prototype GPU clusters primarily for applications of Machine Learning that can be used by university researchers too.