Our research on e-Infrastructure and data engineering ranges from development of new ways to manage large and fast data to massively parallel, interactive and cloud native applications, to management and scalability of cloud and edge infrastructure.
A major part of our work in the lab is aligned with supporting the Cloud 3.0 vision. Apart from conducting original research in distributed systems, we are also working closely with national and international e-Infrastructure providers SNIC and NeIC to develop science cloud IaaS and cloud native AI platforms, with PIs serving in roles ranging from strategic advisors to project managers.
Recent projects has investigated means to better transport, process, store, and visualize large matrices, and other datasets. We study how (and how well) enterprise software such as Apache Spark can be repurposed for scientific computing applications, as well as develop our own frameworks. This work gives rise to optimization problems relating to resource allocation, container placement, and cloud-edge transport. This research finds application in the group’s core experimental work, as well as projects such as HASTE, where microscopy imaging is a key use case.