Lauritz Thamsen | Postdoc @ TU Berlin

I am a postdoctoral researcher at TU Berlin, where I lead the sub-group on Adaptive Resource Management (ARM) in the research group on Distributed and Operating Systems and teach Cloud Computing. I am currently a co-investigator with Odej Kao in the research centers BIFOLD and FONDA, a sub-project lead in WaterGridSense and OPTIMA, a senior researcher in ide3a and HEIBRiDS, as well as an associated researcher with Telemed5000. I also got my PhD from TU Berlin, working on resource management for distributed dataflows in the Berlin Big Data Center and the research project Stratosphere under Odej Kao. Prior to that, I was part of the Software Architecture Group of Robert Hirschfeld at Hasso Plattner Institute, the HPI Research School, and the HPI-Stanford Design Thinking Research Program. I also worked at SAP Labs in Palo Alto on interactive programming systems in SAP's Technology Infrastructure Practice under Dan Ingalls and at Signavio in Berlin as a software engineer.

Lauritz Thamsen

I like distributed systems, operating systems, and software engineering. I am a fan of systems research and generally passionate about making computing more accessible. My specific research interests include resource management, distributed data-parallel processing, and self-adaptive systems. Currently, I am looking a lot at efficient data-parallel processing on diverse cluster infrastructures and dependable distributed processing of sensor streams.

contact

lauritz.thamsen at tu-berlin.de
+49 30 314 24539

office

TEL 1210
TU Berlin

web

twitter: lauritzthamsen
LinkedIn: lauritzthamsen

news

  • July 2020: We are part of FONDA, a new Research Center funded by the DFG. In the project, we will look at how adaptive resource management can make life easier for scientists, who have to work with their available cluster infrastructures to analyze large scientific datasets. If that sounds interesting to you, please take a look at the official announcement of a PhD position in my team. In related news, the open PhD position in the ide3a project is also officially announced now.
  • March 2020: There will be two new and exciting research projects with PhD positions in my team this summer. In BIFOLD, we will continue to do systems research at the intersection of distributed systems, data management, and machine learning. In ide3a, we will set up new teaching around critical urban infrastructures, implementing elements of MOOCs and new simulation tools, in a network of European universities. If that's interesting to you, get in touch!
  • February 2020: I will be staying at the National Institute of Informatics (NII) in Tokyo, Japan, as a visiting researcher in the group of Prof. Fuyuki Ishikawa from April to June this year. The stay is supported by a personal grant from DAAD and I will focus on dependable distributed processing of IoT sensor streams at NII. Update: The stay is postponed until after the coronavirus pandemic.
  • December 2019: We are presenting new tools for efficiently testing potential task placements and system configurations when processing IoT sensor data streams at IEEE/ACM UCC 2019 and IEEE Big Data 2019.
  • November 2019: The research project OPTIMA, in which we investigate and develop a predictive control system for water networks together with partners from Berlin and Brandenburg, was kicked off.
  • October 2019: I am lecturing on cloud computing again at TU Berlin.
  • July 2019: A team of students from our master project won the Atos IT Challenge 2019 with their project idea. The topic of the challenge was to use machine learning for sustainability.
  • April 2019: I am lecturing on distributed systems and co-supervising a master's project on distributed sensor stream processing.
  • December 2018: We are going to present some of our work on scheduling distributed dataflows at IEEE CloudCom 2018 and IEEE Big Data 2018.
  • November 2018: The research project WaterGridSense, in which we develop a scalable analytics platform for continuously processing data from distributed sensors within water networks, was kicked-off at TU Berlin.
  • October 2018: I am lecturing on the methods of cloud computing. I am also co-supervising a bachelor's project on IoT data pipelines and a master's project on applying machine learning for sustainability.
  • May 2018: I successfully defended my PhD thesis on May 4 in front of Odej Kao, Cesar de Rose, Andreas Polze, and Tilmann Rabl. [slides]
  • March 2018: I submitted my PhD thesis on the topic of Dynamic Resource Allocation for Distributed Dataflows to the faculty of Electrical Engineering and Computer Science at TU Berlin. [pdf]
  • December 2017: We are going to present some of our work on runtime prediction and dynamic resource management for batch processing jobs of distributed dataflow systems at IEEE CloudCom 2017 and PDCAT 2017.
  • June 2017: I am going to present some of our work on scheduling recurring batch jobs in shared analytics clusters at IEEE BigData Congress 2017.
  • December 2016: We are presenting our take on runtime prediction for recurring distributed dataflow jobs at IEEE IPCCC 2016 and some work on visual dataflow programming at IEEE Big Data 2016.

availabilities

  • Collaboration: If you are interested in working with me on adaptive resource management in context of distributed data processing, let me know.
  • Community service: I am available for community service in the areas of resource management, distributed data processing, and self-adaptive systems. Please find an up-to-date list of my previous services below.
  • Theses: We are always looking to work with motivated students. Have a look at the proposed topics, at our previous research results, and get in touch.

publications

2020:
  • Mary, Hugo, and Hugo*: Learning to Schedule Distributed Data-Parallel Processing Jobs on Shared Clusters. Lauritz Thamsen, Jossekin Beilharz, Vinh Thuy Tran, Sasho Nedelkoski, and Odej Kao. In Concurrency and Computation: Practice and Experience (e5823). Wiley. 2020. [pdf] [code]
  • Fingerprinting Analog IoT Sensors for Secret-Free Authentication. Felix Lorenz, Lauritz Thamsen, Andreas Wilke, Ilja Behnke, Jens Waldmüller-Littke, Ilya Komarov, Odej Kao, and Manfred Paeschke. To appear in the Workshop Proceedings of the 29th International Conference on Computer Communications and Networks (ICCCN). To be presented at 10th International Workshop on Security, Privacy, Trust, and Machine Learning for Internet of Things (IoTSPT-ML). IEEE. 2020. [preprint]
    • Still to be published. © IEEE, 2020. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists or to reuse any copyrighted component of this work in other works must be obtained from IEEE.
2019:
  • Effectively Testing System Configurations of Critical IoT Analytics Pipelines. Morgan K. Geldenhuys, Lauritz Thamsen, Kain Kordian Gontarska, Felix Lorenz, and Odej Kao. In the Proceedings of the 2019 IEEE International Conference on Big Data (IEEE BigData). Presented at the Second International Workshop on the Internet of Things Data Analytics (IoTDA). IEEE. 2019. [pdf]
    • https://doi.org/BigData47090.2019.9005504, © IEEE, 2019. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists or to reuse any copyrighted component of this work in other works must be obtained from IEEE.
  • Héctor: A Framework for Testing IoT Applications Across Heterogeneous Edge and Cloud Testbeds. Ilja Behnke, Lauritz Thamsen, and Odej Kao. In the Proceedings of the 12th IEEE/ACM International Conference on Utility and Cloud Computing (UCC). Presented at the 8th International Workshop on Cloud and Edge Computing and Applications Management (CloudAM). ACM. 2019. [pdf] [code]
    • © ACM, 2019. This is the authors' version of the work. It is posted here by permission of ACM for your personal use. Not for redistribution. The definitive version will be published in the UCC Companion.
  • Hugo: A Cluster Scheduler that Efficiently Learns to Select Complementary Data-Parallel Jobs. Lauritz Thamsen, Ilya Verbitskiy, Sasho Nedelkoski, Vinh Thuy Tran, Vinícius Meyer, Miguel G. Xavier, Odej Kao, and César A. F. De Rose. In the Proceedings of the Euro-Par 2019 Workshops (Euro-Par). Presented at the 1st International Workshop on Parallel Programming Models in High-Performance Cloud. Springer. 2019. [pdf] [code]
  • Multilayer Active Learning for Efficient Learning and Resource Usage in Distributed IoT Architectures. Sasho Nedelkoski, Lauritz Thamsen, Ilya Verbitskiy, and Odej Kao. In the Proceedings of the 2019 IEEE International Conference on Edge Computing (EDGE). IEEE. 2019. [pdf]
    • https://doi.org/10.1109/EDGE.2019.00015, © IEEE, 2019. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists or to reuse any copyrighted component of this work in other works must be obtained from IEEE.
2018:
  • CoBell: Runtime Prediction for Distributed Dataflow Jobs in Shared Clusters. Ilya Verbitskiy, Lauritz Thamsen, Thomas Renner, and Odej Kao. In the Proceedings of the 10th IEEE International Conference on Cloud Computing Technology and Science (CloudCom). IEEE. 2018. [pdf]
    • https://doi.org/10.1109/CloudCom2018.2018.00029, © IEEE, 2018. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists or to reuse any copyrighted component of this work in other works must be obtained from IEEE.
  • Scheduling Stream Processing Tasks on Geo-Distributed Heterogeneous Resources. Gerrit Janßen, Ilya Verbitskiy, Thomas Renner, and Lauritz Thamsen. In the Proceedings of the 2018 IEEE International Conference on Big Data (IEEE BigData). Presented at the First International Workshop on the Internet of Things Data Analytics (IoTDA). IEEE. 2018. [pdf]
    • https://doi.org/10.1109/BigData.2018.8622651, © IEEE, 2018. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists or to reuse any copyrighted component of this work in other works must be obtained from IEEE.
  • Learning Efficient Co-locations for Scheduling Distributed Dataflows in Shared Clusters. Lauritz Thamsen, Ilya Verbitskiy, Benjamin Rabier, and Odej Kao. In Services Transactions on Big Data (Vol. 4, No. 1). Services Society. 2018. [pdf] [code]
  • Adaptive Resource Management for Distributed Data Analytics. Lauritz Thamsen, Thomas Renner, Ilya Verbitskiy, and Odej Kao. In Lucio Grandinetti, Seyedeh Leili Mirtaheri, Reza Shahbazian, Thomas Sterling, Vladimir Voevodin (eds.), Advances in Parallel Computing – Big Data and HPC: Ecosystem and Convergence. IOS Press. 2018. [pdf]
Previously:
  • Ellis: Dynamically Scaling Distributed Dataflows to Meet Runtime Targets. Lauritz Thamsen, Ilya Verbitskiy, Jossekin Beilharz, Thomas Renner, Andreas Polze, and Odej Kao. In the Proceedings of the 9th IEEE International Conference on Cloud Computing Technology and Science (CloudCom). IEEE. 2017. Best Paper Candidate. [pdf] [code]
    • https://doi.org/10.1109/CloudCom.2017.37, © IEEE, 2017. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists or to reuse any copyrighted component of this work in other works must be obtained from IEEE.
  • SMiPE: Estimating the Progress of Recurring Iterative Distributed Dataflows. Jannis Koch, Lauritz Thamsen, Florian Schmidt, and Odej Kao. In the Proceedings of the 18th International Conference on Parallel and Distributed Computing, Applications and Technologies (PDCAT). IEEE. 2017. [pdf] [code]
    • https://doi.org/10.1109/PDCAT.2017.00034, © IEEE, 2017. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists or to reuse any copyrighted component of this work in other works must be obtained from IEEE.
  • Scheduling Recurring Distributed Dataflow Jobs Based on Resource Utilization and Interference. Lauritz Thamsen, Benjamin Rabier, Florian Schmidt, Thomas Renner, and Odej Kao. In the Proceedings of the 6th IEEE BigData Congress. IEEE. 2017. [pdf] [code]
    • https://doi.org/10.1109/BigDataCongress.2017.28, © IEEE, 2017. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists or to reuse any copyrighted component of this work in other works must be obtained from IEEE.
  • Adaptive Resource Management for Distributed Data Analytics Based on Container-level Cluster Monitoring. Thomas Renner, Lauritz Thamsen, and Odej Kao. In the Proceedings of the 6th International Conference on Data Science, Technology and Applications (DATA). SCITEPRESS. 2017. [pdf] [code]
    • © SCITEPRESS, 2017. This contribution was presented at DATA 17. This is the authors' version of the work.
  • Addressing Hadoop’s Small File Problem With an Appendable Archive File Format. Thomas Renner, Johannes Müller, Lauritz Thamsen, and Odej Kao. In the Proceedings of the Big Data Analytics Workshop (BigDAW), co-located with the ACM International Conference on Computing Frontiers. ACM. 2017. [pdf] [code]
    • © ACM, 2017. This is the authors' version of the work. It is posted here by permission of ACM for your personal use. Not for redistribution. The definitive version will be published in the proceedings of the Big Data Analytics Workshop (BigDAW17).
  • Selecting Resources for Distributed Dataflow Systems According to Runtime Targets. Lauritz Thamsen, Ilya Verbitskiy, Florian Schmidt, Thomas Renner, and Odej Kao. In the Proceedings of the 35th IEEE International Performance Computing and Communications Conference (IPCCC). IEEE. 2016. [pdf] [code]
    • https://doi.org/10.1109/PCCC.2016.7820629, © IEEE, 2016. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists or to reuse any copyrighted component of this work in other works must be obtained from IEEE.
  • CoLoc: Distributed Data and Container Colocation for Data-Intensive Applications. Thomas Renner, Lauritz Thamsen, and Odej Kao. In the Proceedings of the 2016 IEEE International Conference on Big Data (IEEE BigData). Presented at the 4th International Workshop on Distributed Storage Systems and Coding for Big Data. IEEE. 2016. [pdf]
    • https://doi.org/10.1109/BigData.2016.7840954, © IEEE, 2016. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists or to reuse any copyrighted component of this work in other works must be obtained from IEEE.
  • Visually Programming Dataflows for Distributed Data Analytics. Lauritz Thamsen, Thomas Renner, Marvin Byfeld, Markus Paeschke, Daniel Schröder, and Felix Böhm. In the Proceedings of the 2016 IEEE International Conference on Big Data (IEEE BigData). Presented at the 3rd Workshop on Advances in Software and Hardware for Big Data to Knowledge Discovery (ASH). IEEE. 2016. [pdf] [code]
    • https://doi.org/10.1109/BigData.2016.7840860, © IEEE, 2016. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists or to reuse any copyrighted component of this work in other works must be obtained from IEEE.
  • When to Use a Distributed Dataflow Engine: Evaluating the Performance of Apache Flink. Ilya Verbitskiy, Lauritz Thamsen, and Odej Kao. In the Proceedings of the IEEE International Conference on Cloud and Big Data Computing (CBDCom). IEEE. 2016. [pdf]
    • https://doi.org/10.1109/UIC-ATC-ScalCom-CBDCom-IoP-SmartWorld.2016.0114, © IEEE, 2016. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists or to reuse any copyrighted component of this work in other works must be obtained from IEEE.
  • Continuously Improving the Resource Utilization of Iterative Parallel Dataflows. Lauritz Thamsen, Thomas Renner, and Odej Kao. In the Proceedings of the IEEE International Conference on Distributed Computing Systems Workshops (ICDCSW). Presented at the International Workshop on Big Data and Cloud Performance (DCPerf). IEEE. 2016. [pdf] [code]
    • http://dx.doi.org/10.1109/ICDCSW.2016.20, © IEEE, 2016. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists or to reuse any copyrighted component of this work in other works must be obtained from IEEE.
  • Aura: A Flexible Dataflow Engine for Scalable Data Processing. Tobias Herb, Lauritz Thamsen, Thomas Renner, and Odej Kao. In Andreas Knüpfer, Tobias Hilbrich, Christoph Niethammer, José Gracia, Wolfgang E. Nagel, Michael M. Resch (eds.), Tools for High Performance Computing 2015. Springer. 2016. [pdf] [code]
  • Exploratory Authoring of Interactive Content in a Live Environment. Philipp Otto, Jaqueline Pollak, Daniel Werner, Felix Wolff, Bastian Steinert, Lauritz Thamsen, Marcel Taeumel, Jens Lincke, Robert Krahn, Daniel H. H. Ingalls, and Robert Hirschfeld. HPI Technical Reports, vol. 101. Hasso Plattner Institute. 2016. [pdf]
  • Lively Groups: Shared Behavior in a World of Objects without Classes or Prototypes. Tim Felgentreff, Jens Lincke, Robert Hirschfeld, and Lauritz Thamsen. In Proceedings of the Future Programming Workshop (FPW) 2015, co-located with the Conference on Object-oriented Programming, Systems, Languages, and Applications (OOPSLA). ACM. 2015. [pdf]
    • © ACM, 2015. This is the authors' version of the work. It is posted here by permission of ACM for your personal use. Not for redistribution. The definitive version will be published in the proceedings of the Future Programming Workshop (FPW) 2015.
  • Network-Aware Resource Management for Scalable Data Analytics Frameworks. Thomas Renner, Lauritz Thamsen, and Odej Kao. In Proceedings of the First Workshop on Data-Centric Infrastructure for Big Data Science (DIBS) 2015, co-located with the 2015 IEEE International Conference on BigData (BigData). IEEE. 2015. [pdf] [code]
    • http://dx.doi.org/10.1109/BigData.2015.7364083, © IEEE, 2015. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists or to reuse any copyrighted component of this work in other works must be obtained from IEEE.
  • Preserving Access to Previous System States in the Lively Kernel. Lauritz Thamsen, Bastian Steinert, and Robert Hirschfeld. In Hasso Plattner, Christoph Meinel, and Larry Leifer (eds.), Design Thinking Research: Making Design Thinking Foundational. Springer. 2015. [pdf] [code]
  • Implicit Parallelism through Deep Language Embedding. Alexander Alexandrov, Andreas Kunft, Asterios Katsifodimos, Felix Schüler, Lauritz Thamsen, Odej Kao, Tobias Herb, and Volker Markl. In Proceedings of the ACM SIGMOD International Conference on Management of Data (SIGMOD). ACM. 2015. [pdf] [project]
    • © ACM, 2015. This is the authors' version of the work. It is posted here by permission of ACM for your personal use. Not for redistribution. The definitive version will be published in the proceedings of the ACM SIGMOD international conference.
  • Object Versioning to Support Recovery Needs: Using Proxies to Preserve Previous Development States in Lively. Bastian Steinert, Lauritz Thamsen, Tim Felgentreff, and Robert Hirschfeld. In Proceedings of the Dynamic Languages Symposium (DLS) 2014, co-located with the Conference on Object-oriented Programming, Systems, Languages, and Applications (OOPSLA). ACM. 2014. [pdf] [code]
    • © ACM, 2014. This is the authors' version of the work. It is posted here by permission of ACM for your personal use. Not for redistribution. The definitive version will be published in the proceedings of the Dynamic Languages Symposium.
  • Orca: A Single-language Web Framework for Collaborative Development. Lauritz Thamsen, Anton Gulenko, Michael Perscheid, Robert Krahn, Robert Hirschfeld, and David A. Thomas. In Proceedings of the Conference on Creating, Connecting and Collaborating through Computing (C5) 2012. IEEE. 2012. [pdf] [project]
    • http://dx.doi.org/10.1109/C5.2012.9, © IEEE, 2012. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists or to reuse any copyrighted component of this work in other works must be obtained from IEEE.

teaching

  • Lectures: Cloud Computing (Winter 2019 and 2018), Distributed Systems (Summer 2019)
  • Seminars: Hot Topics in Distributed and Operating Systems (Summer 2020), Hot Topics in Distributed System Operations (Winter 2017), Operating Complex IT-Systems (Summer 2015), Advanced Modularity (Winter 2012)
  • Teaching assistance: Systems Programming / Operating Systems (Summer 2018 and 2017), Software Engineering (Winter 2013, 2012, and 2011), Software Architecture (Winter 2010)
  • Supervised projects: Real-Time Applications on IoT Devices (summer 2020), Resource Management for Distributed IoT Stream Processing (Summer 2019), Machine Learning for Sustainability (Winter 2018), Implementing IoT Data Pipelines (Winter 2018), Interactive Cluster Performance Visualization (Winter 2016), Visually Programming Distributed Dataflows (Winter 2015), Software-Defined Networking for Big Data (Winter 2014), Scripting Interactive Web Visualizations (Winter 2013)

community service

  • Program committee: Euro-Par 2020, WWW 2020, COINS 2019, IoTDA 2019, SDNCC 2016
  • External reviewer: IEEE Transactions on Parallel and Distributed Systems, IEEE Transactions on Services Computing, Elsevier's Future Generation Computer Systems, IEEE Access, Springer's Cluster Computing, Wiley's CCPE, ICPE 2020, Euro-Par 2020 workshops, Euro-Par 2019, Euro-Par 2018, IEEE eScience 2018, IEEE ICCAC 2017
  • Student volunteer: AOSD 2012, ESUG 2012

theses

  • Dynamic Resource Allocation for Distributed Dataflows. Lauritz Thamsen. PhD thesis submitted at TU Berlin in March 2018 and successfully defended on May 4, 2018. [pdf] [slides]
  • Object Versioning for the Lively Kernel: Preserving Access to Previous System States in an Object-oriented Programming System. Lauritz Thamsen. Master thesis submitted at Hasso-Plattner-Institut, University of Potsdam, in May 2014. [pdf] [slides] [code]
  • Object Collaboration in the Orca Web Framework. Lauritz Thamsen. Bachelor thesis submitted at Hasso-Plattner-Institut, University of Potsdam, in June 2011. [pdf] [project]

© Lauritz Thamsen   |   Last Update: 09 Aug 2020