Systemy Rozproszone

seminarium magisterskie

rok akademicki 2024/2025

SEMESTR I

• Seminarium 1 - (03.10.24) - Probabilistic DBSCAN for incident detection from support cases (customer tickets) - Kamil Anikiej, Paweł Karwacki, Marcin Wawerka (Snowflake). Sprawy organizacyjne.
• Seminarium 2 - (10.10.24) - Putting an end to accidental SemVer-breaking changes in Rust (Tomasz Nowak), DRust: Language-Guided Distributed Shared Memory with Fine Granularity, Full Transparency, and Ultra Efficiency (Kacper Chętkowski)
• Seminarium 3 - (17.10.24)* - MobileConfig: Remote Configuration Management for Mobile Apps at Hyperscale (Dominik Strąk), Automatically Reasoning About How Systems Code Uses the CPU Cache (Filip Głębocki)
• Seminarium 4 - (24.10.24) - PeRF: Preemption-enabled RDMA Framework (Kacper Sołtysiak), Safe C++ (Jakub Panasiuk)
• Seminarium 5 - (31.10.24) - User-guided Page Merging for Memory Deduplication in Serverless Systems (Hubert Lawenda), An Empirical Study of Rust-for-Linux: The Success, Dissatisfaction, and Compromise (Adam Greloch)
• Seminarium 6 - (07.11.24) - Atomics and Memory Ordering (Szymon Dominikowski)
• Seminarium 7 - (14.11.24) - Serverless HPC: Challenges, Opportunities, and Future Prospects (Marcin Copik, ETH Zurich), Limitations and Opportunities of Modern Hardware Isolation Mechanisms (Franciszek Stachura)
• Seminarium 8 - (21.11.24) - Distributed Ranges - library for distributed and partitioned computations based on C++ ranges (Mikołaj Komar), Harvesting Memory-bound CPU Stall Cycles in Software with MSH (Michał Płachta)
• Seminarium 9 - (28.11.24)* - Sched_ext: pluggable scheduling using BPF (Kernel Recepies on 2023-10-05 by David Vernet, Kernel recepies on 2024-10-01 by David Vernet)
• Seminarium 10 - (05.12.24) - Load is not what you should balance: Introducing Prequal (Szymon Potrzebowski), StreamBox: A Lightweight GPU SandBox for Serverless Inference Workflow (Mikołaj Wasiak) - zdalnie
• Seminarium 11 - (12.12.24)* - Linux Torvalds in conversation with Dirk Hohndel (Open Source Summit Europe on 2024-09-16, Open Source Summit North America on 2024-04-26, Open Source Summit Japan on 2023-12-08)
• Seminarium 12 - (19.12.24) - Memory Barriers: a Hardware View for Software Hackers (Michał Staniewski), Horus: Granular In-Network Task Scheduler for Cloud Datacenters (Artur Kamieniecki)
• Seminarium 13 - (16.01.25) - An exabyte a day: throughput-oriented, large scale, managed data transfers with Effingo (Mateusz Cegiełka), TwinPilots: A New Computing Paradigm for GPU-CPU Parallel LLM Inference (Mateusz Wasilewski)
• Seminarium 14 - (23.01.25) Tematy prac magisterskich Konrad Iwanicki, Paweł Gora, cdn.

SEMESTR II

• Seminarium 1 - (27.02.25) Usage of versioning systems and code review (Tomasz Nowak), SuperBench: Improving Cloud AI Infrastructure Reliability with Proactive Validation (Artur Kamieniecki)
• Seminarium 2 - (06.03.25) Multi-view Feature-based SSD Failure Prediction: What, When, and Why (Hubert Lawenda), HEXO: Offloading long-running compute- and memory-intensive workloads on low-cost, low-power embedded systems (Adam Greloch)
• Seminarium 3 - (13.03.25) Effective implementation of Sparse Matrix as a Data Structure for Distributed Ranges library - praca mgr (Mikołaj Komar)
• Seminarium 4 - (20.03.25) Gasper: Blockchain Consensus Algorithm (Szymon Dominikowski), Scientific Benchmarking of Parallel Computing Systems (Tomasz Kubica)
• Seminarium 5 - (27.03.25) Fast, Transparent Filesystem Microkernel Recovery with Ananke (Michał Płachta), Mechanizmy kontroli dostępu oraz izolacji procesów w Linuxie (Franciszek Stachura)
• Seminarium 6 - (03.04.25) Netcastle: Network Infrastructure Testing At Scale (Szymon Potrzebowski), Combining Buffered I/O and Direct I/O in Distributed File Systems (Dominik Strąk)
• Seminarium 7 - (10.04.25) DeepSeek-V3 Technical Report (Mateusz Cegiełka), Optimizing Scheduler for Linux Gaming (Mateusz Wasilewski)
• Seminarium 8 - (24.04.25) Using Outer Grid in Halo Exchange procedure of Distributed Data Structures in NUMA domain architectures (Kacper Chętkowski), Optimizing the Halo Exchange procedure for the Distributed Ranges library by parallelizing computation and communication (Filip Głębocki)
• Seminarium 9 - (08.05.25) Zajęcia odwołane (Dzień Żałoby na UW)
• Seminarium 10 - (15.05.25) Improving the Performance of Simulating Large Low-Power Wireless Networks in Cooja (Kacper Sołtysiak), "Hey, psst, try this." The underground culture around custom CPU schedulers (Mikołaj Wasiak)
• Seminarium 11 - (22.05.25) SIMD-accelerated regex optimizations (Jakub Panasiuk), Asynchronous dynamic memory allocation on Intel GPUs (Michał Staniewski)
• Seminarium 12 - (29.05.25) Indywidualne rozmowy z magistrantami
• Seminarium 13 - (05.06.25) Indywidualne rozmowy z magistrantami
• Seminarium 14 - (12.06.25) Indywidualne rozmowy z magistrantami