Bio

Gabriel Haas received his B.Sc. in Computer Science from the Technical University of Munich (TUM) in 2016 and his M.Sc. in Software Engineering from the TUM, LMU, and the University of Augsburg in 2019.

He joined the Chair of Decentralized Information Systems and Data Management at TUM in 2022 as a Ph.D. candidate. His research interests include database systems for fast NVMe storage arrays and modern storage technologies.

Publications

Thesis Topics

I currently have the following ideas for theses:

• Implementing and evaluating an LSM-Tree in vmcache.(https://www.cs.cit.tum.de/fileadmin/w00cfj/dis/_my_direct_uploads/vmcache.pdf, https://github.com/viktorleis/vmcache). [Requires strong C++ knowledge.]
• Scalable Snapshot Isolation with thousands of active (user-space) threads/transactions in LeanStore (https://www.vldb.org/pvldb/vol16/p1426-alhomssi.pdf) [Requires strong C++ knowledge.]
• Storage Engine Survey: lots of research papers work on storage engines and key-value stores (e.g.: SplinterDB, TreeLine, LeanStore, Tucana, ...). There would be a lot of value in classifying and categorizing these systems and common ideas. [More like literature review, requires an interest in reading papers and learning about database internals].
• LeanStore Cloud: creating a caching layer for a cloud OLTP database. Similar to Page Servers in Socrates (https://www.microsoft.com/en-us/research/uploads/prod/2019/05/socrates.pdf).

If you are interested in a topic, please read this first www.cs.cit.tum.de/dis/teaching/student-thesis/ and contact me by mail.