2020-11-11, 17:00–17:30, Times in UTC
Cloud computing is a powerful technology that in the last decades revolutionized computing and storage in particular for Industry and Private Sectors. Today, large investments are on-going to build public Cloud Infrastructure at National or International level (e.g. the European Open Science Cloud Initiative). Also, scientists are approaching commercial and public Clouds at different scales: single researchers test the clouds for small research projects, at the same time large international collaborations are evaluating Cloud technology to collect, process, analyze, archive and curate their data.
The experience of large scientific instruments in the last years demonstrates how experiments are critically dependent on computing, data processing and storage infrastructures and our ability to utilise them through codes and algorithms. This will become more and more important with the advent of new class instruments (e.g. SKA, CTA, LSST, EUCLID) where new computational challenges will be faced. This will lead to the use of new infrastructures where exascale HPC and Clouds will converge to answer new challenges of (Big-) data analysis and (Big-) data analytics (HPDA).
New technologies (e.g. containerization) are driving the convergence of these “worlds” and the advent of science platforms as mean to access data, storage and computing (but also software and algorithms) is facilitating the use of cloud at different scales.
In this talk, I will discuss the actual use of Cloud in Astrophysics at different scale using some examples and I will present future trends and possibilities that the use of cloud computing and its convergence with HPC will open: from HTC to HPDA, from scientific computing to data analytics.