Research Programs

Systems and Networks

Mission

This research laboratory is concerned with the design and implementation of modern distributed systems and networks, increasingly characterized by stringent requirements in terms of high performance, Quality of Service, large scale, dynamicity, and security.

The research laboratory tackles the challenges of distributed systems at all levels, spanning from the definition of systems, to distributed algorithms, to middleware and language constructs, to the implementation of application-level protocols and system services (including the ones provided by operating system kernel, the drivers, but also user-space system daemons that are critical for the system's reliability and performance, and for the QoS perceived by the users), to routing and networking problems, including P2P paradigm and wireless systems (WSN, mesh networks, ad-hoc systems, ...).

A strong emphasis is placed on validating the research outcomes through implementation of real systems and simulations.

Research Areas

System Support

The Operating System (including kernel, drivers, user-space components and libraries, etc...) is particularly important for the whole distributes system's performance.
In order to provide high performance and a predictable QoS to distributed applications, we research kernel support, high-performance device drivers and network stacks, virtualisation technologies (as a support for cloud computing), and novel OS functionalities for systems ranging from small embedded devices to large multi-core servers.
Our focus is on open-source architectures, that simplify the implementation of novel research ideas (thanks to reuse a large open-source codebase), and allow to easily make them available to the users in a short time.

Middleware

We design programming models meeting the needs of modern distributed systems and the corresponding distributed runtime, focusing on an efficient implementation in the face of dynamicity and large scale.

Application-level protocols

Modern distributed systems rely more and more on an application-level substrate ensuring efficient communication. Examples are protocols for content-based routing in publish-subscribe, and for building and maintaining overlay networks. Along with traditional deterministic protocols, we also study the applicability of probabilistic, gossip-based algorithms to a large number of functionalities in overlay networks, ranking from aggregation, load balancing, resource assignments, topology construction and maintenance, etc.

Distributed algorithms

We develop novel techniques for an efficient and effective distribution and replication of data objects in a way that is transparent to end users and applications. Special emphasis is given to a number of properties such as quality, fault-tolerance, reliability, dependability and consistency, all for data of large scales.

Security & privacy

We design and develop security architectures and protocols to support and provide security, trust and privacy properties in very large distributed systems. One recurrent challenge is to meet security requirements within the stringent constraints (i.e. performance, usability, scalability) of the target system. We are also interested in the study of new access control models for emerging distributed applications as social networks.

Networking

We study and develop network protocols and architectures at all levels, starting from MAC protocols, to routing, to congestion control, to P2P overlay systems, to vehicular networks. Performance-oriented design is one of the key focus of our research, as well as cross-layer design in those cases (mobile networks, heterogeneous overlays, ...) when strict black-box layering cannot yield optimized performance.

Applications Areas

Large-scale distributed computing and peer-to-peer networks

Several of our application-level protocols are applied in the implementation of systems and tools for very large-scale distributed systems, on both wired and wireless settings.

Cloud Computing and Virtual Networks

We investigate how to implement high-performance cloud computing platforms providing predictable QoS to the users. This can be obtained by applying new OS concepts and real-time technologies to virtualisation.
The Cloud Computing idea can also be extended to implement Virtual Networks. In particular, we are currently working on virtual software routers, improving their predictability and performance.

Wireless sensor networks

We have expertise in designing and implement various elements of the software layer of wireless sensor networks, including system-level mechanisms, routing protocols, as well as middleware platforms for simplifying the programming chore. Moreover, we have extensive expertise in deploying the hw/sw system for the needs of real-world applications.

Mobile, context-aware computing

We research the application of SOA in mobile environment, with particular focus on traffic applications and local/contextual human-participated service provisioning, where humans can provide local, real-time services to people in the same area (or in the same context).

P2P streaming and mutimedia support

The design of Video and Voice streaming systems is one of the applications that encompasses several expertise. Beyond traditional architectures based on IP systems, P2P-based streaming is becoming commercially interesting and important, form P2P-TV to video conferencing, to on-demand content distribution.

Traffic and Road Safety 

As parte of the Inter Vehicular Networking we are particularly interested in applications to improve safety and road utilization. This includes emergency braking, platooning, collision avoidance in crossroads and so forth. We are among the key developers of Veins, one of the most advanced and diffused platforms for the integrated study of mobility and communications, and the only one that allows the study of safety applications including the evaluation of the accidents severity. 

Wireless Community Networks

Bottom-up broadband networking is one of the emerging paradigms to overcome the architectural limitations of today Internet, and not only a last-mile replacement for rural areas. They can be seen as a low-level P2P application if you want, but indeed they are a field where trans-disciplinary approaches (from economy, sociology, law, ...) are needed to find the proper socio-technical architecture that can support the development and management of the network itself as a parallel player (not substitute) of the Internet. 

Members

Faculty Members

Renato Lo CignoAlberto Montresor (Coord)Gian Pietro Picco
Yannis Velegrakis

Further Information

Technical reports of Research Program
Published papers of Research Program

Project

Jan 01, 2016 netCommons network infrastructure as commons