Dr. Felix Martin Schuhknecht

Postdoctoral Researcher

Overview

Dr. Felix Martin Schuhknecht
Postdoctoral Researcher
Big Data Analytics Group
Campus E1.1, room 3.09
66123 Saarbrücken, Germany

e-mail: felix.schuhknecht at bigdata.uni-saarland.de
phone: +49 (0) 681 302 70146
github: https://github.com/FMSchuhknecht

I am a Postdoctoral Researcher at the Saarland University's Computer Science Department. In 2016, I finished by Ph.D. under the supervision of Prof. Dr. Jens Dittrich. I am working in the field of blockchains, transactional management, adaptive indexing, sorting, partitioning, and memory management. Besides, I am one of the co-founders of "ChainifyDB" (www.chainifydb.com).

In November 2020, I will start at the Johannes-Gutenberg University Mainz as a professor. If you are interested in a PhD or Postdoc position in the area of information systems, data management, or blockchain systems, feel free to contact me.

Felix Martin Schuhknecht

Activities

  1. PC Member / Journal Reviewer
  2. (External) Reviewer
  3. Funding
  4. Misc

Publications

Research Projects

  • Blockchain systems.
    • ChainifyDB:
      • A blockchain layer that can be installed on top of established data management infractuctures to equip them with trust.
      • Funded with 840.000€ by BMBF.
      • Goal: Develop market-ready product.
    • Transitioning database technology to Hyperledger Fabric:
      • We apply mature database technology in form of transaction reordering and early transaction abort to the permissioned blockchain system Hyperledger Fabric.
      • This drastically improves the throughput of successful transactions, that flow through the system.
  • Interleaving of database management system and operating system.
    • RUMA / Memory Rewiring:
      • A method to manipulate the mapping from virtual to physical main memory at runtime from user space in a vanilla Linux kernel.
      • We apply RUMA to increase the performance of various data structures and algorithms in the context of database management systems.
    • AnKer:
      • Enables efficient, concurrent processing of transactional and analytical queries.
      • Short-running transactions are fully serializable via multi-version concurrency control (MVCC), while long-running analytical transactions operate on a separate virtual memory area.
      • AnKer utilizes custom system calls to enable extremely fast virtual snapshotting. This further enhances the interleaving of DBMS and OS.
  • Adaptively indexing large datasets.
    • Analysis and improvement of adaptive indexing techniques with respect to convergence speed, robustness against workload patterns, and efficient tuple reconstruction.
    • Development of a "meta-adaptive index", that is able to emulate a large number of adaptive indexes.
    • Development of concurrent adaptive indexing methods, utilizing state-of-the-art hardware features.
    • Analysis and improvement of partitioning and sorting methods, utilizing state-of-the-art hardware features.
  • Application of machine learning techniques to enhance database manegement systems.
    • We apply deep reinforcement learning to automatically configure a DBMS (e.g. creation of indices, adjustment of buffer sizes) with respect to a workload.
  • System architecture, physical layouts, and compression.
    • Integration of a column-oriented storage layout into a row-ortiented DBMS (IBM DB2) via user defined functions (UDFs)
    • Integration of various compression techniques to the introduced column-oriented storage layout.
    • Analysis of different relational systems (PostgreSQL, MonetDB, HyPer) and NoSQL systems (Redis, Aerospike) with respect to processing performance under typical workloads for these system classes.

Teaching

Courses

SS 2018 Proseminar "13 Wochen - 13 Blockchains" (5 CP)
SS 2018 Seminar "Current Topics in BigData Management" (7 CP)
WS 2017/2018 Chief teaching assistant of the core lecture “Database Systems” (9 CP)
SS 2017 Chief teaching assistant of the proseminar “13 Wochen – 13 Datenbanksysteme” (5 CP)
SS 2017 Teaching assistant of the seminar “Current Topics in BigData Management” (7 CP)
WS 2016/2017 Teaching assistant of the seminar “Current Topics in BigData Management” (7 CP)
WS 2015/2016 Chief teaching assistant of the core lecture “Database Systems” (9 CP)
WS 2014/2015 Teaching assistant of the seminar “Database Systems 2” (7 CP)
SS 2014 Chief teaching assistant of the base lecture “Information Systems” (6 CP)
WS 2013/2014 Teaching assistant of the seminar “Current Topics in Big Data Management” (7 CP)
SS 2013 Teaching assistant of the seminar “Current Research in Data Management” (7 CP)
SS 2011 Teaching assistant of the base lecture “Information Systems” (6 CP)

Supervised Students

Sep. 2016 – Nov. 2017 Master thesis of David Gembalczyk
Nov. 2015 – Jul. 2016 Bachelor thesis of David Gembalczyk
Apr. 2015 – May 2016 Master thesis of Tobias Frey
Apr. 2015 – Apr. 2016 Master thesis of Laurent Linden
Mar. 2014 – May 2014 Master practical training of Pankaj Khanchandani (6 CP)
May 2013 – May 2014 Master thesis of Stefan Tombers

Positions

  • March 2020 - September 2020
    Principle Investigator
    BMBF project "ChainifyDB", Saarland University

  • January 2017 - September 2020
    Postdoctoral Researcher
    Big Data Analytics Group, Saarland University

  • January 2013 - December 2016
    Ph.D. Student
    Big Data Analytics Group, Saarland University

Studies

  • January 2013 - December 2016
    Ph.D. in Computer Science (summa cum laude)
    Saarland University

  • November 2010 - January 2013
    M.Sc. in Computer Science
    Saarland University

  • October 2007 - November 2010
    B.Sc. in Computer Science
    Saarland University