Parallel

Computing

and I/O

Michael Blesel has joined our group as a new doctoral researcher. Welcome!

Michael gave his inaugural lecture on “Parallel Computing and I/O - Herausforderungen des datenintensiven Hochleistungsrechnens” via Zoom.

Michael gave a talk about “Converging Storage Technologies Using a Flexible HPC Storage Framework” at CHPC National Conference 2020.

Our workshop Challenges and Opportunities of Efficient and Performant Storage Systems (CHEOPS) has been accepted at EuroSys 2021 and will take place on April 26, 2021. Michael and Kira will organize it together with colleagues from DataDirect Networks and ENSTA Bretagne.

Michael and Kira's paper “Coupling Storage Systems and Self-Describing Data Formats for Global Metadata Management” has been accepted at CSCI 2020.

We offer a lecture on parallel programming this semester.

Michael presented the paper Improving Energy Efficiency of Scientific Data Compression with Decision Trees at ENERGY 2020. He was also part of the panel on “Energy Data and Adaptive Consumption”, where he gave a talk about “Improving Energy Efficiency in High Performance Computing by Powering Down Unused Resources”. Moreover, our student Daniel presented the paper ArduPower v2: Open and Modular Power Measurement for HPC Components.

Together with colleagues from Sandia National Laboratories and Intel, Michael and Kira organized the Challenges and Opportunities of HPC Storage Systems (CHAOSS) workshop at Euro-Par 2020.

It is aimed at researchers, developers of scientific applications, engineers and everyone interested in the evolution of HPC storage systems. As the developments of computing power, storage and network technologies continue to diverge, the performance gap between them widens. This trend, combined with the growing data volumes, results in I/O and storage bottlenecks that become increasingly serious especially for large-scale HPC storage systems. The hierarchy of different storage technologies to ease this situation leads to a complex environment which will become even more challenging for future exascale systems.