Nikolas Ioannou

prof_pic.jpg

Google Scholar
GitHub
LinkedIn
Twitter

Nikolas Ioannou is an Engineering Manager and Technical Lead at YouTube (Google), leading the ML Efficiency SRE group. He currently oversees the tracking and improvement of utilization efficiency across the YouTube ML training and serving fleets. Previously, Nikolas was the TLM for YouTube Metadata SRE (2022–2025), working on the infrastructure serving YouTube’s metadata to more than 1.5 billion users per month.

Before Google, Nikolas was a Research Staff Member & Manager, and Master Inventor at IBM Research Zurich. There, he worked on cloud infrastructure, non-volatile memories, distributed file systems, and systems aspects of machine learning frameworks. His work delivered a 50% cost reduction in critical IBM Cloud Database services and a 2x performance improvement to IBM Cloud serverless and Spark services. His research has impacted a range of IBM products, including IBM FlashSystem, IBM Easy Tier Server, and IBM Watson Machine Learning.

Nikolas holds more than 120 patents and has authored over 50 articles in journals such as ACM ToS, ACM TACO, IEEE TPDS, and IBM JRD, and in major conferences including Usenix FAST, VLDB, NeurIPS, MICRO, and PACT. He has received a best paper award (SYSTOR) and two spotlight awards (NeurIPS, MICCAI). He is an ACM Distinguished Speaker (2021–2024, 2024–2027) and a Senior Member. Nikolas received his ECE MEng degree from the National Technical University of Athens and his PhD in Computer Science from the University of Edinburgh.

selected publications

  1. Understanding modern storage APIs: a systematic study of libaio, SPDK, and io_uring
    Diego Didona, Jonas Pfefferle, Nikolas Ioannou, Bernard Metzler, and Animesh Trivedi
    SYSTOR ’22: The 15th ACM International Systems and Storage Conference, Haifa, Israel, June 13 - 15, 2022
  2. Towards a General Framework for ML-based Self-tuning Databases
    Thomas Schmied, Diego Didona, Andreas C. Döring, Thomas P. Parnell, and Nikolas Ioannou
    EuroMLSys@EuroSys 2021, Proceedings of the 1st Workshop on Machine Learning and Systemsg Virtual Event, Edinburgh, Scotland, UK, 26 April, 2021
  3. Toward a Better Understanding and Evaluation of Tree Structures on Flash SSDs
    Diego Didona, Nikolas Ioannou, Radu Stoica, and Kornilios Kourtis
    Proc. VLDB Endow., 2020
  4. SnapBoost: A Heterogeneous Boosting Machine
    Thomas P. Parnell, Andreea Anghel, Malgorzata Lazuka, Nikolas Ioannou, Sebastian Kurella, Peshal Agarwal, Nikolaos Papandreou, and Haralampos Pozidis
    Advances in Neural Information Processing Systems 33: Annual Conference on Neural Information Processing Systems 2020, NeurIPS 2020, December 6-12, 2020, virtual
  5. Reaping the performance of fast NVM storage with uDepot
    Kornilios Kourtis, Nikolas Ioannou, and Ioannis Koltsidas
    17th USENIX Conference on File and Storage Technologies, FAST 2019, Boston, MA, February 25-28, 2019
  6. Understanding the Design Trade-Offs of Hybrid Flash Controllers
    Radu Stoica, Roman A. Pletka, Nikolas Ioannou, Nikolaos Papandreou, Sasa Tomic, and Haris Pozidis
    27th IEEE International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems, MASCOTS 2019, Rennes, France, October 21-25, 2019
  7. Accelerated ML-Assisted Tumor Detection in High-Resolution Histopathology Images
    Nikolas Ioannou, Milos Stanisavljevic, Andreea Anghel, Nikolaos Papandreou, Sonali Andani, Jan Hendrik Rüschoff, Peter Wild, Maria Gabrani, and Haralampos Pozidis
    Medical Image Computing and Computer Assisted Intervention - MICCAI 2019 - 22nd International Conference, Shenzhen, China, October 13-17, 2019, Proceedings, Part I
    Spotlight paper award
  8. SySCD: A System-Aware Parallel Coordinate Descent Algorithm
    Nikolas Ioannou, Celestine Mendler-Dünner, and Thomas P. Parnell
    Advances in Neural Information Processing Systems 32: Annual Conference on Neural Information Processing Systems 2019, NeurIPS 2019, December 8-14, 2019, Vancouver, BC, Canada
    Spotlight paper award
  9. Management of Next-Generation NAND Flash to Achieve Enterprise-Level Endurance and Latency Targets
    Roman A. Pletka, Ioannis Koltsidas, Nikolas Ioannou, Sasa Tomic, Nikolaos Papandreou, Thomas P. Parnell, Haralampos Pozidis, Aaron Fry, and Tim Fisher
    ACM Trans. Storage, 2018
  10. Elevating Commodity Storage with the SALSA Host Translation Layer
    Nikolas Ioannou, Kornilios Kourtis, and Ioannis Koltsidas
    26th IEEE International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems, MASCOTS 2018, Milwaukee, WI, USA, September 25-28, 2018
    Best paper candidate
  11. Phase-Based Application-Driven Hierarchical Power Management on the Single-chip Cloud Computer
    Nikolas Ioannou, Michael Kauschke, Matthias Gries, and Marcelo Cintra
    2011 International Conference on Parallel Architectures and Compilation Techniques, PACT 2011, Galveston, TX, USA, October 10-14, 2011
    Best paper candidate