Philipp Wiesner

I’m a research associate and 4th year PhD student at TU Berlin in the Distributed and Operating Systems (DOS) group by Prof. Odej Kao. Before that, I’ve worked as a Software and Data Engineer at different companies in Berlin.

My research focus is on carbon-aware computing: Aligning the power consumption of computing systems with the availability of renewable energy. Besides that, I’m interested in systems for machine learning and the co-simulation of computing and energy systems.

Find me on Google Scholar , GitHub , LinkedIn , and Twitter or contact me at wiesner at tu-berlin.de

News

October 2024: I’ll present our latest research on the trade-off between carbon footprint and user experience of ML inference at his month’s BIFOLD Lunch Talk.
September 2024: Our paper Federated Learning over Connected Modes was accepted at NeurIPS'24.
August 2024: We’re organizing the 1st International Workshop on Low Carbon Computing (LOCO 2024) to take place on December 3, 2024 in Glasgow and online.
July 2024: My Software Campus project SYNERGY is coming to an end after 2.5 years. I’ve been leading a research team on sustaibale federated learning (FedZero, e-Energy'24) and the testing of carbon-aware systems (Vessim, HotCarbon'24).
June 2024: Our paper Vessim: A Testbed for Carbon-Aware Applications and Systems has been accepted at HotCarbon'24.

more ...

April 2024: The recording of my latest talk on carbon-efficient federated learning at the Flower Monthly April is now online.
March 2024: Check out our new blog article on sustainable FL and my poster presentation at the Flower AI Summit 2024.
Januar 2024: I am visiting Prof César De Rose at PUCRS in Porto Alegre, Brasil for a 6 week research stay.
November 2023: Our paper FedZero: Leveraging Renewable Excess Energy in Federated Learning was accepted as a full paper at ACM e-Energy'24 which had an 8% acceptance rate (4/47) at their fall deadline.
September 2023: Dennis Grinwald and I will be giving a talk on efficient ML at the 2nd BIFOLD Weizenbaum summer school.
August 2023: Our paper Software-in-the-Loop Simulation for Developing and Testing Carbon-Aware Applications was accepted for publication in Software: Practice and Experience.
July 2023: We released Vessim in GitHub, the first co-simulation testbed for carbon-aware applications and systems.
June 2023: I’ll present our research on sustainable federated learning at the Software Campus Summit 2023.
April 2023: We have three new student workers starting at DOS this month, growing our team on the Synergy project to five people.
March 2023: I presented or current research on leveraging renewable excess energy in federated fearning at the Low-Carbon and Sustainable Computing seminar series organized by the University of Glasgow.
January 2023: I’m visiting the IRIT in Toulouse for two weeks to collaborate on their DATAZERO2 project.
November 2022: Our solution Lowcarb won the 1st price in CarbonHack22 organized by the Green Software Foundation.
October 2022: This winter semester I’ll supervise our Master Project: Distributed Systems, which will be the first course on Carbon-Aware Computing at TU Berlin.
August 2022: We presented Cucumber at Euro-Par 2022 in Glasgow.
July 2022: I have been invited to a discussion on carbon-aware computing in the Environment Variables podcast of the Green Software Foundation. Listen to the episode here!
June 2022: Our paper Cucumber has been featured in a BIFOLD blog article and a press release by TU Berlin.
May 2022: We’re organizing the 2nd iteration of our workshop on Testing Distributed IoT Systems (TDIS), submit your work by June 21!
March 2022: I will be visiting Lauritz Thamsen and Dimitrios Pezaros for a research stay at the Glasgow Systems Section of the Univesity of Glasgow for four months.
February 2022: This month my Software Campus project on synergies between distributed power generation and edge intelligence starts in collaboration wiuth Huawei. I’ll be leading this project over the next two years.

Selected Publications

Vessim: A Testbed for Carbon-Aware Applications and Systems
Philipp Wiesner, Ilja Behnke, Paul Kilian, Marvin Steinke, and Odej Kao
3rd Workshop on Sustainable Computer Systems (HotCarbon). ACM. 2024.
Workshop preprint code video BIFOLD news
FedZero: Leveraging Renewable Excess Energy in Federated Learning
Philipp Wiesner, Ramin Khalili, Dennis Grinwald, Pratik Agrawal, Lauritz Thamsen, and Odej Kao
15th ACM International Conference on Future and Sustainable Energy Systems (e-Energy). ACM. 2024.
Conference preprint code video BIFOLD news
Let's Wait Awhile: How Temporal Workload Shifting Can Reduce Carbon Emissions in the Cloud
Philipp Wiesner, Ilja Behnke, Dominik Scheinert, Kordian Gontarska, and Lauritz Thamsen
22nd ACM/IFIP International Middleware Conference (Middleware). ACM. 2021.
Conference preprint code video BIFOLD news; TU Berlin press release; New Scientist article; hackernews frontpage