Delay Tolerant Networks (DTN) are networks of self-organizing wireless nodes, where end-to-end connectivity is intermittent. In these networks, content or information between nodes is exchanged (opportunistically), whenever two nodes are within range ("in contact"). Forwarding decisions are generally probabilistic and based on locally collected knowledge about node behavior (e.g., past contacts between nodes) to predict future contact opportunities. The use of complex network analysis has been recently suggested to perform this prediction task and improve the performance of opportunistic (DTN) routing. Contacts seen in the past are aggregated to a "Social Graph", and a variety of metrics (e.g., entrality and similarity) or algorithms (e.g., community detection) can be used to assess the utility of a node to deliver a content or bring it closer to the destination.

The current Internet includes a large number of distributed services. In order to guarantee the QoS of the communication in these services, a client has to select a close-by server with enough available resources. In order to achieve this objective, in this Thesis, we propose a simple and practical solution for Dynamic and Location Aware Server Discovery based on a Distributed Hash Table (DHT). Specifically, we decide to use a Chord DHT system (although any other DHT scheme can be used). In more detail, the solution works as follows. The servers offering a given service form a Chord-like DHT. In addition, they register their location (topological and/or geographical) information in the DHT. Each client using the service is connected to at least one server from the DHT. Eventually, a given client realizes that it is connected to a server providing a bad QoS, then, it queries the DHT in order to find an appropriate server (i.e. a close-by server with enough available resources). We define 11 design criteria, and compare our solution to the State of the Art based on them. We show that our solution is the most complete one. Furthermore, we validate the performance of our solution in two different scenarios: NAT Traversal Server Discovery and Home Agent Discovery in Mobile IP scenarios. The former serves to validate our solution in a highly dynamic environment whereas the latter demonstrates the appropriateness of our solution in more classical environments where the servers are typically hosts.

La conferencia se impartirá en inglés

Wireless ad hoc networks have emerged to be used in scenarios where it is required or desired to have wireless communications among a variety of devices without relying on any infrastructure or central management. One of the fundamental operations in wireless ad hoc networks is broadcasting, where a wireless device (simply called a node) disseminates a message to all other nodes in the network. A major challenge of efficient broadcast algorithms is to reduce the number of transmissions required to disseminate a message. Unfortunately, minimizing the total number of required transmissions is an NP-hard problem even when the whole network topology is known by every node. Reducing the number of transmissions becomes more challenging in local broadcast algorithms, where each node makes decision (whether or not to transmit a received message) based on local neighborhood information. The common belief is that local broadcast algorithms are not able to guarantee both full delivery and a good bound on the number of transmissions.

In this talk I present a platform to extract models of security-relevant functionality from program binaries, enabling multiple security applications such as active botnet infiltration, finding deviations between implementations of the same functionality, vulnerability signature generation, and finding content-sniffing cross-site scripting (XSS) attacks. In this talk, I present two applications: active botnet infiltration and finding content-sniffing XSS attacks.

His talk consists out of two parts. The first part will give an overview on the current research activities and achievements of the Distributed Multimedia Systems Research Group at the University of Oslo. This includes video streaming in MANETs and disruptive environments, publish subscribe for sparse MANETs, deviation detection with complex event processing for automated home care systems, and clean-slate Future Internet research work. The second part will focus on CacheCast, which is joint work with Lancaster University and has been initiated in the Content NoE. Due to the lack of multicast services in the Internet, applications based on single source multiple destinations transfers such as video conferencing, IP radio, IPTV must use unicast or application layer multicast. 

Combined advances in high speed networking, mobile devices, application sharing, web services, virtual world technologies and large scale event processing are converging to create a new world of pervasive, ubiquitous “presence” of users, which offers tremendous potential for social interaction and co-creation. The communication networking and computing requirements of this converged human-centric environment are also increasing at an accelerated pace. In this new environment, it is imperative that the much-needed networking and computing resources align closely with the needs and patterns dictated by the applications, social networks, and by the human users. We believe that the success of such socio-technical systems will hinge on the way networks capture and interact with human presence and location, in all of its physical, virtual and perceived aspects.

Nowadays it is possible to watch some TV channels in the Internet using P2P mechanisms. There are basically two ways to construct a P2P network to transmit streaming video: mesh-based and tree-based (there is also a mix of both called hybrid-based). No matter how we construct the overlay, a fundamental problem is: what is the best peer to connect with, in order to obtain the best performance? This question is even more important in the tree-based P2P networks where there is just one connection between a "parent" peer and each one of its children, so, the leaving of a parent forces all its children to reconnect to other parents. Our goal in this study is to minimize the number of orphan peers per minute, i.e. select a parent who minimizes the probability of leaving before a given peer. In order to do that, it is necessary to have other results like the distribution of the channel holding time per peer, the future lifetime of a peer given its elapsed time in a given channel and other stuff that will be described during the presentation.

Existing cloud computing platforms offer virtually unlimited compute resources (virtual machines, bandwidth, storage, etc.) that can be used on demand. Such on-demand model offers significant elasticity to the customers in terms when and where they use the resources. The existing pricing model, however, is pay-as-you-go which in turn can lead to unpredictable costs to the cloud customers. This talk will discuss two adaptive approaches for resource control under a fixed budget: Distributed Rate Limiting (DRL) and Temporal Rate Limiting (TRL). DRL is a fully decentralized mechanism for resource control over a distributed cloud service, that splits the available budget among the participating nodes subject to the load each node experiences. TRL in contrast, splits the budget over a time period, to optimize the performance of the customer with demand pattern that varies in time.

Existing cloud computing platforms offer virtually unlimited compute resources (virtual machines, bandwidth, storage, etc.) that can be used on demand. Such on-demand model offers significant elasticity to the customers in terms when and where they use the resources. The existing pricing model, however, is pay-as-you-go which in turn can lead to unpredictable costs to the cloud customers. This talk will discuss two adaptive approaches for resource control under a fixed budget: Distributed Rate Limiting (DRL) and Temporal Rate Limiting (TRL). DRL is a fully decentralized mechanism for resource control over a distributed cloud service, that splits the available budget among the participating nodes subject to the load each node experiences. TRL in contrast, splits the budget over a time period, to optimize the performance of the customer with demand pattern that varies in time.