Wikipedia's content is being written by a completely decentralized set of volunteers across the Globe. However, the centralized server architecture does not match the decentralization of content contribution. We investigate how one could hypothetically redesign the entire Wikipedia system architecture to allow volunteers to host the content collaboratively and securely.
- Wikipedia Workload Analysis for Decentralized Hosting, Guido Urdaneta, Guillaume Pierre and Maarten van Steen, Computer Networks 53(11), July 2009.
- A Survey of DHT Security Techniques, Guido Urdaneta, Guillaume Pierre and Maarten van Steen. ACM Computing Surveys, to appear.
- More information...
Dependable Cloud computing
The Cloud computing project investigates the way to build scalable and consistent Web applications in the Cloud, and to guarantee constant user-perceived performance even though the request workload may vary by several orders of magnitude and the environment conditions may change. We employ techniques such as scalable NoSQL data services and resource provisioning.
- Service-Oriented Data Denormalization for Scalable Web Applications, Zhou Wei, Jiang Dejun, Guillaume Pierre, Chi-Hung Chi and Maarten van Steen. In Proc. WWW, April 2008.
- Autonomous Resource Provisioning for Multi-Service Web Application, Jiang Dejun, Guillaume Pierre and Chi-Huing Chi. In Proc. WWW, April 2010.
- CloudTPS: Scalable Transactions for Web Applications in the Cloud, Zhou Wei, Guillaume Pierre and Chi-Hung Chi. Technical report, February 2010.
- More information here and there.
Extreme-scale ad-hoc networks
Wireless sensor networks connect together hundreds to thousands of tiny batery-operated sensor nodes communicating by wireless transmission. They are good examples of spatial distributed systems where a node's physical location determines what it can and cannot do. We study ad-hoc wireless networks that are extreme in their size (up to dozens of thousands of nodes) and their resource constraints (limited energy, feeble ratio and processing power). We use gossip-based protocols to devise creative application-level scenarios and to improve the energy efficiency of the MAC layer.
- Gossip-Based Self Management of a Recursive Area Hierarchy for Large Wireless SensorNets, Konrad Iwanicki and Maarten van Steen, IEEE Transactions on Parallel and Distributed Systems 21(4), April 2010.
- On Hierarchical Routing in Wireless Sensor Networks, Konrad Iwanicki and Maarten van Steen, In Proc. IPSN, April 2009.
- More publications on this topic
The XtreemOS E.U. project is building an open source Grid Operating System with native support for virtual organizations and capable of running on a wide range of underlying platforms: PCs, clusters, mobile devices, and grids. Our group has been working on core technologies to allow scalable operation of the Grid: the Resource Selection Service implements the fundamental operation that selects a set of resources that are suitable to execute a given operation. Distributed servers implement a network-level service to provide distributed services with location-transparency.
- Autonomous Resource Selection for Decentralized Utility Computing, Paolo Costa, Jeff Napper, Guillaume Pierre and Maarten van Steen, In Proc. ICDCS, June 2009.
- Adam2: Reliable Distribution Estimation in Decentralised Environments, Jan Sacha, Jeff Napper, Corina Stratan and Guillaume Pierre, In Proc. ICDCS, June 2010.
- Enabling Service Adaptability with Versatile Anycast, Michal Szymaniak, Guillaume Pierre, Mariana Simons-Nikilova and Maarten van Steen, Concurrency and Computation: Practice and Experience 19(3), Sept. 2007.
- More information...
You may also want to take a look at our previous research projects.