Gian Paolo Jesi

Proximity-aware superpeer overlay topologies

The concept of superpeer has been introduced to improve the performance of popular P2P applications. A superpeer is a “powerful” node that acts as a server for a set of clients, and as an equal with respect to other superpeers. By exploiting heterogeneity, the superpeer paradigm can lead to improved efficiency, without compromising the decentralized nature of P2P networks. The main issues in the construction of superpeer-based overlays are the selection of superpeers, and the association between superpeers and clients. Generally, superpeers are either run voluntarily (without an explicit selection process), or chosen among the “best” nodes in the network, for example those with the most abondant resources, such as bandwidth or storage. In several contexts, however, shared resources are not the only factor; latency between clients and superpeers may play an important role, for example in online games. This paper presents SG-2, a novel protocol for building and maintaining a proximity-aware superpeer topology. SG-2 uses a gossip-based protocol to spread messages to nearby nodes and a biology-inspired task allocation mechanism to promote the “best” nodes to superpeer status. The paper includes extensive simulation experiments to prove the efficiency, scalability and robustness of SG-2.

Secure peer sampling

Gossiping has been identified as a useful building block for the development of large-scale, decentralized collaborative systems. With gossiping, individual nodes periodically interact with random partners, exchanging information about their local state; yet, they may globally provide several useful services, such as information diffusion, topology management, monitoring, load-balancing, etc. One fundamental building block for developing gossip protocols is peer sampling, which provides nodes with the ability to sample the entire population of nodes in order to randomly select a gossip partner. In existing implementations, however, one fundamental aspect is neglected: security. Byzantine nodes may subvert the peer sampling service and bias the random selection process, for example, by increasing the probability that a fellow malicious node is selected instead of a random one. The contribution of this paper is an extension to existing peer sampling protocols with a detection mechanism that identifies and blacklists nodes that are suspected of behaving maliciously. An extensive experimental evaluation shows that our extension is efficient in dealing with a large number of malicious nodes.

Building Latency-aware Overlay Topologies with QuickPee

This work presents a gossip-based protocol, termed QuickPeer, which builds and maintains latency-aware overlay topologies. Such topologies are useful for several distributed applications, like distributed online gaming, context-aware P2P applications and QoS-aware publish/subscribe systems. The distinctive feature of QuickPeer is that it can manage large scale overlay topologies providing each host in the overlay with its closest or furthest neighbor, according to network distance (RTT), in few gossip rounds. We present experimental results that prove that QuickPeer is a scalable and robust solution for large-scale latency-aware overlay topology management

Canning spam in wireless gossip networks

Once a problem associated only with email, spam is now affecting other media, such as instant messaging, blogs, newsgroups and mobile phone messaging. As wireless networks become more commonplace, we can expect that spam will find its way into upcoming wireless communication services. This paper studies the threat posed by malicious nodes inserting spam in a wireless network using gossiping as a method for information dissemination. We identify the security mechanisms needed to protect our gossip network against the proliferation of spam, reducing the problem to a matter of finding and removing corrupted messages. Finally, we propose a probabilistic method of integrity checking to contain the spread of spam which we evaluate through extensive simulations

Organizational Learning with Agent Based Modeling

In this work, we adopt simulation in order to explore organization learning in assembling operations as an emergent phenomenon. We focus on the emergence of organizational learning curves which is supposed to be the signal that leads to the spontaneous arise of routines or patterns of operations. In particular, we focus on pure routing problems which are sufficient for the generation of learning curves. We carry out our study using our own developed simulation framework - ACP - and we show how it has been straightforward to develop such kind of simulation model using the features available in the framework.