Distributed Systems Group
The Distributed Systems Group at INESC-ID (the computer science and engineering research laboratory associated with IST – University of Lisbon) conducts research in several topics in the broad area of computer systems. The overarching goal of our research is to improve the abstractions, mechanisms and techniques that are used for building distributed and concurrent applications, so that these applications become more robust, more secure, and perform better.
Ad hoc and wireless computing. This area addresses the research challenges that arise in the context of ad hoc and wireless networks, such as sensor networks, wireless meshes and IoT.
Autonomic computing. This research line investigates the design of pervasive self-tuning mechanisms that aim to maximize the efficiency and robustness of complex, large scale, parallel and distributed systems. The methodologies employed to achieve these goals range from analytical models, to machine learning, control theory and simulation models.
Large-scale parallel computations. We are studying better ways to perform large-scale batch or stream processing by leveraging, modifying, and extending existing “big data” infrastructures.
Scaling up reliable systems. We are devising methods and protocols that enable robust distributed systems to scale in several dimensions, namely: in terms of the size of the service state and correspondingly the number of servers that store it; in terms of the number of clients that simultaneously access a shared part of that state; or in terms of the number of geographic locations where that state must be stored in order to ensure a low latency access to end users.
Security. We are working on the design and implementation of secure distributed systems. Our research focuses on system mechanisms and policies for the implementation of secure distributed systems, such as authentication, communication streams and filesystems.
Transaction processing. This topic addresses research issues such as transactional support for distributed applications, database replication, and transactional memory.
Our research is supported by: