Ilin Tolovski

Ph.D. Student
Room: F-1.04
Phone: + 49- (0) 331 5509-4885
E-Mail: Ilin.Tolovski(at)hpi.de

Full profile

Research Interests

• Machine Learning Systems
• Interoperability in Collaborative Environments
• Efficiency in Machine Learning Pipelines
• Benchmarking of ML Systems
• Sustainability of Data Processing Systems

Publications

• Green or Greedy? An Ecological Analysis of Data Center GPU Replacements @ICDE '26
  Marc Baeuerle, Ole Becker, Nikolas Hoellerl, Ricardo Salazar-Díaz, Ilin Tolovski, Tilmann Rabl
  > Paper
• TPCx-AI under the Microscope: A Benchmarking Debt Analysis @VLDB '26
  Ilin Tolovski, Philipp Hildebrandt, Khuzaima Daudjee, Tilmann Rabl
  > Paper
• Efficiently Joining Large Relations on Multi-GPU Systems @VLDB '25
  Tobias Maltenberger, Ilin Tolovski, Tilmann Rabl
  > Paper
• TCO2: Analyzing the Carbon Footprint of Database Server Replacements @VLDB '25
  Marc Baeuerle, Thomas Bodner, Martin Boissier, Tilmann Rabl, Ricardo Salazar-Díaz, Florian Schmeller, Nils Strassenburg, Ilin Tolovski, Marcel Weisgut, Wang Yue
  > Paper
• A Case for Ecological Efficiency in Database Server Lifecycles @CIDR '25
  Thomas Bodner, Martin Boissier, Tilmann Rabl, Ricardo Salazar-Díaz, Florian Schmeller, Nils Strassenburg, Ilin Tolovski, Marcel Weisgut, Wang Yue
  > Paper
• Addressing Data Management Challenges for Interoperable Data Science @DATAI '24 (co-located with VLDB)
  Ilin Tolovski, Tilmann Rabl
  > Paper
• RMG Sort: Radix-Partitioning-Based Multi-GPU Sorting @BTW '23
  Ivan Ilic, Ilin Tolovski, Tilmann Rabl
  > Paper
• Efficiently Managing Deep Learning Models in a Distributed Environment @EDBT '22
  Nils Strassenburg, Ilin Tolovski, Tilmann Rabl
  > Paper
• Evaluating Multi-GPU Sorting with Modern Interconnects @SIGMOD '22
  Tobias Maltenberger, Ivan Ilic, Ilin Tolovski, Tilmann Rabl
  > Paper
• DAPHNE: An Open and Extensible System Infrastructure for Integrated Data Analysis Pipelines
  @ CIDR '22 
  Pattrick Damme, Marius Birkenbach, Constantinos Bitsakos, Matthias Boehm, Phillipe Bonnet, Florina Ciorba, Mark Dokter, Pawl Dowgiallo, Ahmed Eleliemy, Christian Faerber, Georgios Goumas, Dirk Habich, Niclas Hedam, Marlies Hofer, Wenjun Huang, Kevin Innerebner, Vasileios Karakostas, Roman Kern, Tomaž Kosar, Alexander Krause, Daniel Krems, Andreas Laber, Wolfgang Lehner, Eric Mier, Tilmann Rabl, Piotr Ratuszniak, Pedro Silva, Nikolai Skupin, Andreas Starzacher, Benjamin Steinwender, Ilin Tolovski, Pinar Tözün, Wojciech Ulatowski, Yuanyuan Wang, Izajasz Wrosz, Aleš Zamuda, Ce Zhang, Xiao Xiang Zhu
  > Paper
• A Survey of Big Data, High Performance Computing, and Machine Learning Benchmarks
  @ TPCTC '21 (co-located with VLDB)
  Nina Ihde, Paula Marten, Ahmed Eleliemy, Gabrielle Poerwawinata, Pedro Silva, Ilin Tolovski, Florina M. Ciorba, Tilmann Rabl
  > Paper

Teaching

Lectures:

• Big Data Lab (Master, Winter 2023/24)
• Big Data Systems (Master, Winter 2020/21, 2021/22)
• Database Systems I (Bachelor, Summer 2020)
• Database Systems II (Bachelor, Winter 2025/26)
• Hardware-conscious Data Processing (Master, Summer 2022, 2023, 2024, 2025)

Projects/Seminars:

• Sustainability in Data Management (Master, Summer 2025)
• Machine Learning Systems (Master, Winter 2023/24)
• Interoperability in Data Processing Pipelines (Master, Winter 2022/23)
• Data Management in ML Systems (Master, Winter 2021/22)
• GPU-Accelerated Data Processing (Master, Winter 2020/21)
• Data Processing on Modern Hardware (Master, Summer 2020)

Master & Bachelor Projects:

• Tracking Carbon: Multi-Tier Analysis of Compute Job Emissions in Datacenters (Bachelor Project, 2023/24)
• Model Lifecycle Analysis Platform (Master Project, Summer 2022)
• End-to-end ML System Benchmarking (Bachelor Project, 2020/21)

Thesis Supervision: 

If you are interested in writing your thesis with me, have a look at the open theses page for some ideas and guidelines. If you have an interesting idea that you want to work on, feel free to drop me a line.

Completed Theses:

Master Theses:

• Squirrel: A Carbon-Aware HPC Scheduler utilizing Spatiotemporal Workload Shifting, Luca Springer, 2024
• Efficiently Joining Large Relations on Multi-GPU Systems with Modern Interconnects, Tobias Maltenberger, 2024
• RMG Sort: Radix-Partitioning-Based Multi-GPU Sort, Ivan Ilic, 2022
• Model Management in Distributed Environments, Nils Straßenburg, 2021

Bachelor Theses:

• Data Center Energy Consumption: Dataset Publication and Analysis, Alexander Probst, 2026
• A Monitoring Solution for Energy Data in Data Centers, Alexander Netz, 2024
• Integrating and Designing a Database Schema for the HPI Data Center’s Energy Data, Theo Klinke, 2024
• Benchmarking Distributed Machine Learning Systems, Willi Rieck, 2021
• Benchmarking Containerized Machine Learning Pipelines, Jost Morgenstern, 2021
• Real-Time Benchmarking, Jonas Schulze, 2021
• UI Designs for Benchmarking Systems, Christian Jacob, 2021