93468 - Distributed Software Systems

Learning outcomes

The course presents some current technologies to develop distributed systems. The concept of distributed system is introduced first, then the concept of middleware technology as a distributed programming platform. The fundamental principles of distributed systems will be explained. This introductory material is followed by a description of the main architectures of distributed systems, including those for implementing and deploying service-based applications in the cloud. At the end of the course the students will be able to select the appropriate middleware platform for a given problem and to apply the main technologies presented to solve practical case studies. The exam will consist of 1) a midterm presentation on a research paper chosen by the student and 2) a final project assigned by the instructor.

Course contents

This course introduces the fundamental principles of distributed software systems, with a focus on communication, concurrency, coordination, fault tolerance, and middleware-based design. It combines theory and practice to help students design and reason about reliable distributed applications, while highlighting where LLMs can be used as high-value components in modern distributed architectures.

Syllabus: Introduction to distributed systems; distributed programming principles; middleware technologies; time, ordering, and consensus; transparency, replication, and failure handling; cloud, edge, and service-oriented architectures; practical labs on distributed application design, including selected LLM-enabled use cases such as orchestration, service integration, and system observability

Readings/Bibliography

Distributed Systems (4th ed., 2023) by van Steen & Tanenbaum, available online.

Additional papers concerning distributed systems and LLM as components will be distributed 

Teaching methods

Short interactive lectures, guided discussion, in-class problem solving, and hands-on labs. A flipped-classroom style will be used for selected topics, with pre-reading and paper discussion before class. Labs and mini-projects will focus on designing, implementing, and evaluating distributed services. Selected sessions will include case studies on LLM-enabled distributed architectures, with emphasis on design trade-offs, reliability, and observability.

Assessment methods

The assessment consists of laboratory assignments, a midterm report and presentation, and a final exam. Labs evaluate practical skills, the report and presentation assess the ability to analyze and communicate a distributed-systems topic, and the final exam checks understanding of the core theoretical concepts. Selected LLM-related aspects may be included in labs or the project, especially for architecture, reliability, and observability.

Teaching tools

Lecture slides, lab assignments, reference papers, and a reproducible coding environment. Students will use a shared code repository, container-based setups, and short demo notebooks for experiments. Selected sessions may include LLM-based tools for observability, debugging, or service orchestration, but only as supporting material for the core distributed-systems content.

We will use Virtuale and MS Teams.

Office hours

See the website of Paolo Ciancarini