Personal Data

My research interests concentrate on large-scale distributed systems, in particular those exhibithing a high degree of dynamicity. These include peer-to-peer networks, whereby the physical network is fixed but bears at its top a logical network known as overlay network. Reconfiguration occurs at logical level and is triggered by the massive number of peers joining and leaving the network. Another example are wireless networks such as sensor and vehicular ad hoc networks in which devices can come and go continuously, e.g.,because they are free to move since connected through wireless links or because they can fail at any time and new devices can get in. In these reconfigurable scenarios, communication and coordination of components is often a challenging task, given the unreliability of the environment, including the communication network. Middleware technologies have the ability to offer a simplified view of the network and the environment to applications, addressing heterogeneity, distribution, and resource sharing as well as increasing the dependability of the applications by building applications on a solid, verified base. Middleware supporting mobility (both physical and logical) have to react to frequent changes in the environment due to the movement of mobile users or to node unexpected connections and disconnections. Unfortunately, traditional platforms targetting group communication do not address dynamic reconfig-uration, thus hampering their exploitation in such scenarios. The focus of my research is therefore on the development of decentralized protocols, mechanisms, and ultimately implementations that support large scale and highly dynamic environments, including peer-to-peer networks, sensor networks and mobile ad hoc networks. Hereafter I concisely illustrate my research approaches and results, integrated by an essential bibliographic reference 1. Content-based routing (CBR) differs from classical routing paradigms as messages are routed based on their content rather than their destination address. This form of implicit, multi-point communication fosters a high degree of decoupling, since the communicating parties are not necessarily aware of each other, and can therefore change dynamically without affecting the rest of the system. These ideas are at the core of several implemented systems, including publish-subscribe and event notification systems, distributed databases, peer-to-peer applications, and data collection in wireless sensor networks. In these middleware layers, the programmer is provided with considerable expressive power, by enabling the data consumer (the subscriber) to dynamically specify the filtering of relevant data based on its content, instead of letting the source (the publisher) bind it explicitly to some static notion of group or topic, …