Distributed systems are inherently event-based systems communicating over a network. They are challenging to design and implement as they introduce fundamental problems related to latency, concurrency and partial failures, as pointed out by Waldo in the seminal article “A note on distributed computing”. To support the development of distributed systems, we argue that it is essential to provide novel programming abstractions which are explored in tandem with software tools tailored to modern concurrent and distributed software.

In this talk, we first discuss examples of programming language abstraction to support the implementation and verification of distributed systems, focusing on the challenges of data replication and consistency. We then turn our attention to concrete examples to ease debugging and the difficult task of finding the root cause of bugs in distributed systems. In particular, we discuss examples of debugging techniques to deal with the event-based nature of those systems, focusing on recent work on debugging IoT systems.

Bio:

Elisa Gonzalez Boix is a professor at the Vrije Universiteit Brussel (VUB). She leads the Distributed and Concurrency (DisCo) group within the Software Languages Lab of the VUB, where she and her students investigate language technology to support the design and implementation of programming abstractions and tools for modern concurrent and distributed systems. Her later works focus on web-based distributed systems, geo-replicated systems and the Internet of Things. She publishes in programming and software engineering venues like ECOOP, OOPSLA, and Programming, and actively contributes to organising events in her area, like the PaPoc and DEBT workshops and the DARE summer school. She received her PhD in Sciences at the VUB and her bachelor’s and master’s degrees from Universitat Politècnica de Catalunya (UPC). You can find Elisa on Twitter as @elisagboix and find information on her research at: https://soft.vub.ac.be/disco