The lab is teaching the following courses:

• Concurrent Computing (CS-453) (theory & practice)
• Distributed Algorithms (CS-451) (theory & practice)

The lab taught in the past the following courses:

• Information, Calcul et Communication
• Introduction à la Programmation Orientée Objet
• Secure Distributed Computing (Fall semester 2009)

DCL offers master projects in the following areas:

• Cryptocurrencies: We have several project openings as part of our ongoing research on designing new cryptocurrency systems. Please contact Prof. Rachid Guerraoui.

• Scalable Distributed Cache Coherency: The distributed cache coherency problem is a critical challenge in modern computing systems, especially as cloud platforms, large-scale web services, and in-memory data grids become increasingly central to industry. In distributed systems, ensuring that all nodes have a consistent view of shared data, even when copies are cached locally, is essential for correctness and reliability. Without coherent caches, stale reads can lead to data races, inconsistency, and subtle bugs—issues that can severely impact services like Amazon DynamoDB, Google Spanner, or Meta’s TAO graph store, where distributed caching is used at massive scale. Companies like Google and Microsoft invest heavily in research and infrastructure (e.g., Tardis, Azure Service Fabric) to implement scalable, low-latency cache coherence protocols across thousands of machines. Efficient distributed cache coherence not only improves performance and availability but also enables stronger consistency guarantees in systems that underpin everything from real-time recommendations to financial transactions. As systems grow more distributed and memory-centric, solving this problem becomes foundational to scaling modern software infrastructures. However, current distributed cache coherency protocols suffer from excessive communication overhead, limiting scalability. The question is how to offset such an overhead while maintaining the overall protocol's performance. If interested, contact Beatrice Shokry.

• Accelerating Safe ML Systems: ML has been a hot topic for so long. Now with LLMs, it is getting even more attractive for everyone in the research community as well as industry (e.g., Google, Meta, etc.). In particular, training large models with massive data makes the need for distributed computing (i.e., distributing tasks among machines) non questionable, which leads to two main challenges. First, how to do it fast? Second, how to do it safe (e.g., secure collaborative training, robust ML, etc.)? At the heart of these two challenges is how to communicate with other machines in a fast and a secure way? This leads us to Remote Direct Memory Access (RDMA) technology which is becoming increasingly important in the field of machine learning (ML), particularly for distributed training of large models and handling massive datasets. RDMA enables high-throughput, low-latency data transfers between servers without involving the CPU, which significantly reduces the overhead associated with traditional networking methods. This is crucial for ML tasks that require rapid synchronization and communication among multiple nodes. Now the question is how to use RDMA efficiently to build fast and secure ML systems? If interested, contact Beatrice Shokry for more information.

• Distributed Resource Efficient Inference Engine: The software stack for running efficient Generative AI inference is tuned for data centers, optimized for uniform, well connected hardware monitored by expert teams. While this is perfect for hyperscalars, many private companies and public institutions of every sizes have raising concerns about privacy and sovereignty, and yet, are unable to deploy their own GenAI software stack. The DCL and its recent spin-off Anyway have developed a distributed inference engine, called “Anyd” to address these issues. Anyd aggregates hardware from different generations and different vendors, connected by any network, into a unified cluster that automatically decides how to use these resources to run inference and handles failures transparently. Projects of this category aim to address the specific challenges that come with the distributed and automatically managed nature of Anyd. Contact Geovani Rizk for more information.

If the subject of a Master Project interests you as a Semester Project, please contact the supervisor of the Master Project to see if it can be considered for a Semester Project.

EPFL I&C duration, credits and workload information are available on https://www.epfl.ch/schools/ic/education/master/semester-project-msc/.