General Regulations

• Type of course: Master Seminar, Summer Term 20116
• Teaching Staff: Dr. Matthias Uflacker, Martin Boissier, Markus Dreseler
• Location: Villa, 2.16, Hasso-Plattner-High-Tech-Park, August-Bebel-Str. 88
• The seminar will mostly be held in individual meetings and not during the assigned time slot.
• First Meeting: April 11 at 1:30 PM, D-E.9/10
• In-Memory Technology Deep Dive: April 18 at 1:30 PM, D-E.9/10
• Topic Selections due: April 19
• Even though the seminar has already started, you can still join until April 19.
• Introductory Slides
• Final Presentation: tbd.
• 4 semester periods per week (Semesterwochenstunden)
• 6 credit points (graded)
• Area of specialization: BPET, OSIS, SAMT

Short Description

The goal of the research seminar is to teach the students the basics of scientific research and a basic knowledge of the inner mechanics of in-memory databases. The seminar is focused on recent research topics as well as latest hardware trends (e.g., non-volatile memory, 240 core server machines, co-processing) in conjunction with columnar in-memory database systems.

Each student will work individually on a topic, resulting in a final paper (8-10 pages, IEEE) in addition to a midterm and final presentation. The topics range from low-level implementations of database structures to applications that are made possible thanks to in-memory technology and will be defined together with the students. In the first meeting, we will propose more detailed topics.

All topics have an implementation component, which should be employed and closely evaluated in the resulting paper.

Results of previous offerings of this seminar have been published at different venues (ADC '14, IMDM '15, IOTBD '16) and have been part of continued research. In this semester, the following topics will be covered:

• Data Stream Processing with Kafka Streams
• Recognizing Compound in Events in Spatio-Temporal Football Data
• Hyrise-R: Heterogeneous replicas and quick integration of new replicas
• Using Infiniband to Speed Up Networking
• Hot and Cold Data
• GPU Computing for Enterprise Applications

Find more information in the introductory slides.

The following components determine the final mark:

Presentations (Mid-term / Final): 30 (10 / 20) %

Results: 30 %

Article: 30 %

Peer-Review of an assigned paper of a fellow student: 10 %

All of the components must be passed in order to pass the seminar.