Workshop: Reducing carbon emissions in large-scale computational data analysis

On November 25th and 26th, FONDA will host a workshop on reducing carbon emissions in large-scale computational data analysis with our co-organizers from the Quantitative Biology Center (QBiC), the bioinformatics core facility from the University of Tübingen.

Researchers from across Europe and the UK have been invited to present their research on monitoring and predicting the carbon footprint of large scale computational data analysis. This ongoing work promotes sustainability by providing information about how to reduce the carbon emissions of large scale data analysis and scientific workflows.

Location

Humboldt Universität zu Berlin, Institute for Informatics:

Humboldt-Kabinett (1st floor seminar room)

Johann-von-Neumann Haus

Rudower Chaussee 25

12489, Berlin

Schedule

Tuesday, Nov. 25th

TimeEventSpeakerTitle
12:00-13:00Lunch (open to all)  
13:00-14:00Internal meetings, Guests arrive  
14:00-14:05WelcomeUlf Leser 
14:05-15:05 Philipp Wiesner -Technische Universität BerlinThe Sustainability Risks of Reasoning AI
Loïc Lannelongue – University of CambridgeGreen Algorithms, Green DiSC and GREENER principles: making computational science more environmentally sustainable (online)
15:05-15:30Break  
15:30-16:30 Lauritz Thamsen – University of GlasgowCarbon-Conscious Scalable Data Analysis
w/ the Ichnos Carbon Footprint Estimator
(online)
Philipp Thamm (HU-Berlin) – FONDA subproject A2  Energy Attribution for Parallel Workflow Tasks in Nextflow on Kubernetes Clusters
16:30-17:00Break  
17:00-17:30 Nadja Volkmann and Josua Carl – Quantitative Biology Center in Tübingennf-co2footprint: A Nextflow plugin to estimate the carbon footprint of pipeline runs
18:00Dinner (self-paid)  

Wednesday, Nov 26th

TimeEventSpeakerTitle
9:00-9:30Arrival, welcome  
9:30-10:30 Norbert Eicker – Jülich Supercomputing Centre & University of WuppertalSEANERGYS: towards a European software stack for energy efficient operation of HPC/AI centres
Anita Ragyanszki – Zuse Institute Berlin Sustainable Supercomputing: Operational Strategies for Reduced Power Consumption
10:30-11:00Break  
11:00-12:00 Manuel Marciani – Barcelona Supercomputer Center Overview of energy measurement in leading-class HPC centers
Ilin Tolovski (HPI) and Matthias Boehm (TU-Berlin) – FONDA subproject B6 Energy-aware Profiling and Optimization of ML Pipelines
12:00-13:00Lunch  
13:00-14:00 Sebastian Tiesler – IZ GreenComputeSupporting Energy Efficient Data Analysis at Humboldt University
Mirko Schäfer – NFDI4EnergyEnergy system data for granular emission accounting of electricity consumption
14:00-14:30Break  
14:30-15:30 Dominik Schweisgut (KIT) – FONDA subproject B1Carbon-Aware Workflow Scheduling with Deadline Constraint
Kathleen West – University of GlasgowCarbon-Aware Scientific Workflow Execution
15:30-16:00Wrap up, Farewell  

Icons made by Vectors Tank and juicy_fish from www.flaticon.com