FID3218 Advanced Topics in Data Systems 7.5 credits

The course features ~30 seminar sessions during the academic year (four periods), typically held weekly. Each seminar includes a research presentation and a moderated discussion. Students participate in presentations, moderation, and active discussions, drawing on topics from their own research, invited talks, and recent papers from venues such as CIDR, VLDB, SIGMOD, SOSP, OSDI, and EuroSys. The concrete topics of the seminars are:

• Cloud‑native and serverless data architectures; custom hardware acceleration, memory hierarchies and networking for data systems .

• Query processing, indexing, transaction processing and concurrency control at scale .

• Data quality, integration, provenance, semantics and responsible data management .

• Streaming, complex event processing and approximate/uncertain databases .

• Systems and algorithms for AI/ML workloads and ML for data management .

• Operating systems, virtualization and runtime support for data‑intensive workloads .

• File, storage and caching systems; distributed and edge/fog systems; reliable and secure system design .

• Research visions from CIDR and cross‑disciplinary topics that explore novel hardware or emerging paradigms such as blockchains, serverless analytics and data systems for spatial/time‑series data.

On completion, students will be able to:

1. Analyze advanced research directions in scalable data systems and systems‑level support.

2. Critically evaluate data management research papers, identifying contributions, methodology, and limitations.

3. Analyze research results (their own or from the literature) to an expert audience clearly and persuasively.

4. Moderate and synthesize technical discussions, providing constructive feedback.

5. Synthesize and critically evaluate data-systems research in relation to broader systems topics such as operating systems, distributed platforms, and hardware.

6. Create and articulate a forward-looking research direction in data systems, integrating insights from multiple seminar topics and identifying emerging challenges and opportunities.

The course is recommended for students that have already completed ID2203 or equivalent course FID3011. This is not a hard requirement but a recommendation since the seminars will assume that the doctoral students have a basic understanding of distributed data consistency as well as distributed coordination problems which are prevalent in cloud-based data systems today.

• EXA1 - Examination, 7.5 credits, grading scale: P, F

Based on recommendation from KTH’s coordinator for disabilities, the examiner will decide how to adapt an examination for students with documented disability. The examiner may apply another examination format when re-examining individual students. If the course is discontinued, students may request to be examined during the following two academic years.

Paris Carbone

• All members of a group are responsible for the group's work.
• In any assessment, every student shall honestly disclose any help received and sources used.
• In an oral assessment, every student shall be able to present and answer questions about the entire assignment and solution.