|Service talk: Scientific Writing |
|Submit concept (via
ISIS)||by 17.05. 23:59h||PDFs|
|Presentation dates (three per date)||10.6., 17.6, 24.6, 1.7,|
|Submit report (via ISIS)||by
ISIS and Registration
- There is an ISIS course for the organization of this seminar (for all announcements, slides, discussions etc.): https://isis.tu-berlin.de/course/view.php?id=24301 .
- Registration for this seminar is required
(capacity: 12) and will be possible through the
- The possibility to register opens on 12.04.2021 and closes 18.04.2021.
- Successful registrations will be confirmed on 19.04.2021.
- At the same time, you will be required to submit your topic preferences.
Scalable & Reliable Distributed Systems for Data-Intensive Applications
In this seminar we will take a look at important classes of scalable and fault-tolerant distributed systems that are commonly used in context of data-intensive applications.
Today, many organizations have to deal with very large volumes of collected data, be it to recommend songs or TV shows to millions of users, to accurately identify genetic disorders by comparing terabytes of human genomic data, to monitor current environment conditions such as seismic activities using large-scale distributed sensor networks, or to detect anomalies or fraudulent behavior in millions of business transactions. For this, businesses, sciences, municipalities, as well as other large and small organizations deploy data-intensive applications that run on scalable and reliable distributed systems and on large-scale computing infrastructures. Prominent examples of such systems include distributed file systems, dataflow systems, distributed messaging systems, and libraries for scalable machine learning as well as graph processing.
Based on individual preferences, the participants of the seminar are assigned a particular class of scalable systems, then prepare a presentation and an essay on that topic, describing the motivation, basic concepts, major developments, as well as current research around their class of scalable systems. Participants are expected to both consult textbooks and research papers and to take a look at actual systems.
Specific system classes of interest may include:
- messaging systems,
- distributed storage systems,
- scalable databases,
- cluster resource managers,
- distributed data processing systems,
- distributed stream processing systems,
- systems for machine learning,
- graph processing systems,
- scientific workflow systems,
- serverless computing platforms,
- monitoring systems
Students may also propose own topics, if they fit the theme of this seminar. The final list of topics will be made available with the kickoff.
Master students MSc-Inf and MSc-TI, also Erasmus computer science students and business informatics (MSc) students.
- Independently investigate, comprehend and reflect a class of important computer systems
- Present the topic orally (20 minutes presentation + 10 minutes discussion)
- Write a scientific report (max. 6 pages, IEEE Template for Conference Proceedings )
Literature and Other Materials
- Designing Data-Intensive Applications, Kleppmann, 2017
- Designing Distributed Systems: Patterns and Paradigms for Scalable, Reliable Services, Burns, 2017
- Distributed Systems: Principles and Paradigms, van Steen and Tanenbaum, any ed.
Recommendations for additional literature and other materials will be given based on the individual topics.
This seminar fulfills the module "Master Seminar: Operating Complex IT-Systems " (3 LP, portfolio examination, presentation and report).
+49 30 314-24539
e-mail query 
+49 30 314-79675
Room TEL 1211
e-mail query