86. Towards cloud-native scientific computing

Bartosz Balis– Institute of Computer Science and ACC Cyfronet, AGH, Krakow, PL

Abstract

Cloud technologies have been influencing the scientific community for at least a decade. Virtualization, containerization, or object storage services are now routinely used in HPC data centers. What will come next? In this talk, we’ll give an introduction to Kubernetes, a cornerstone cloud-native technology, and explore its potential for scientific computing management. We’ll also look at other examples of leveraging cloud computing frameworks in scientific computing, including an application of Google Dataproc for tightly coupled  large-scale parallel processing.

About the author

Bartosz Baliś (PhD, DSc) currently works as an associate professor at the Institute of Computer Science AGH and a senior researcher at Sano. He is also involved in the ALICE experiment in CERN. His research interests include scientific workflow management, cloud computing, cloud-native scientific computing, big data processing and data engineering. He has developed HyperFlow, a cloud-native scientific workflow management system. He has participated in national and EU-funded research projects CrossGrid, CoreGRID, K-Wf Grid, ViroLab, Gredia, PL-Grid, UrbanFlood, ISMOP, PaaSage, and WATERLINE.