One Computer Science MSc thesis to be presented on 12 June

Monday, 12 June there will be a master thesis presentation in Computer Science at Lund University, Faculty of Engineering.

The presentation will take place in room E:2405 (Glasburen).

Note to potential opponents: (Register as an opponent to the presentation of your choice by sending an email to the examiner for that presentation (firstname.lastname@cs.lth.se). Do not forget to specify the presentation you register for! Note that the number of opponents may be limited (often to two), so you might be forced to choose another presentation if you register too late. Registrations are individual, just as the oppositions are! More instructions are found on this page.)

---

09:15-10:00 in E:2405 (Glasburen)

Presenters: Lisa af Klint, Evelina Danielsson
Title: Predicting Loss of Fault Tolerance in a Cloud Graph Database
Examiner: Markus Borg
Supervisors: Johan Eker (LTH), Olof Englund (Neo4j)

With applications which are hosted on the cloud becoming popular, many systems adapt fault tolerance to be able to continue operating without loss of functionality when one or more of its components fail. Every anomalous behaviour in a system is a potential cause to an incident that can lead to a system losing this quality. This thesis investigates if monitoring metrics can be used to alert when a database cluster is at risk of losing fault tolerance. After identifying interesting metrics, two different machine learning models were implemented and evaluated in its success of identifying anomalies correlating to fault intolerance. The authors found that the only metric that showed a reasonable correlation was haproxy.backend.errors.con_rate. Additionally, neither model was successful in predicting when a database was at risk of becoming fault intolerant. The models used were HBOS and LSTM-based autoencoder, where HBOS performed marginally better.

Link to popular science summary: https://fileadmin.cs.lth.se/cs/Education/Examensarbete/Popsci/230612_09DanielssonFranzenafKlint.pdf