Section: Other Grants and Activities
Google Research Award “CRDTs: Consistency without concurrency control”
A CRDT is a data type whose operations commute when they are concurrent. Replicas of a CRDT eventually converge without any complex concurrency control or the need for any centralised component. This makes CRDTs very appealing for managing data in large-scale environments, such as cloud computing or web-based environments. We have previously developed two non-trivial CRDTs: a shared edit buffer, Treedoc, and a graph structure, the multilog. This work allowed us to identify some general properties for the design of CRDTs  ,  . The goal of this work is to generalise this approach to manage data in large-scale environments, by designing CRDTs for specific problems (e.g. replicated key-value store as used in Amazon Dynamo).
The principal investigators of this award are Marc Shapiro and Nuno Preguiça of UNL. This award includes a grant of $80 000 over one year.
Grant from Microsoft Research Cambridge
Data replication enables cooperative work, improves access latency to data shared through the network, and improves availability in the presence of failures. This grant supports a doctoral student for studying consistency between replicas of mutable, semantically-rich data in a peer-to-peer fashion. This study should enable to engineer distributed systems and applications based on them, supporting cooperative applications in large-scale collaboration networks. It includes a systematic exploration of the solution space, in order to expose the cost vs. performance vs. availability vs. quality trade-offs, and understanding fault tolerance and recovery aspects. This work combines formal approaches, simulation, implementation, and measurement.
Grid4All - (2006-2009)
France Télécom Recherche et Développement, INRIA (Regal, Atlas and Grand-Large), SICS, KTH, ICCS, UPRC, UPC, Redidia.
European Commission, 6th Framework Programme, STREP (Specific Targeted Research Project)
Grid4All embraces the vision of a “democratic” Grid as a ubiquitous utility whereby domestic users, small organizations and enterprises may draw on resources on the Internet without having to individually invest and manage computing and IT resources. This project is funded by the 6th Framework Programme of the European Commission. It involves institutional and industrial partners. Its budget is slightly over 4.8 million euros.
Grid4All has the following objectives:
To alleviate administration and management of large scale distributed IT infrastructure, by pioneering the application of component based management architectures to self-organizing peer-to-peer overlay services.
To provide self-management capabilities, to improve scalability, resilience to failures and volatility thus paving the way to mature solutions enabling deployment of Grids on the wide Internet.
To widen the scope of Grid technologies by enabling on-demand creation and maintenance of dynamically evolving scalable virtual organisations even short lived.
To apply advanced application frameworks for collaborative data sharing applications executing in dynamic environments.
To capitalize on Grids as revenue generating sources to implement utility models of computing but using resources on the Internet.
Grid4All will help to bring global computing to the broader society beyond that of academia and large enterprises by providing an opportunity to small organisations and individuals to reap the cost benefit of resource sharing without however the burdens of management, security, and administration.
The consortium will demonstrate this by applying Grid4All in two different application domains: collaborative tools for e-learning targeting schools and digital content processing applications targeting residential users.
FTH-GRID - (2009–2010)
Université de Lisbonne (LASIGE), LIP6 (Regal)
FTH-Grid, Fault-Tolerant Hierarchical Grid Scheduling, is a cooperation project between the Laboratoire d'Informatique de Paris 6 (LIP6/CNRS, France) and the Large-Scale Informatics Systems Laboratory (LASIGE/FCUL, Portugal).
Its goal is to foster scientific research collaboration between the two research teams. The project aims at rendering Map Reduce on top of Grid tolerant to byzantine failure. Map Reduce is a programming model for large-scale data-parallel applications whose implementation is based on master-slave scheduling of bag-of-tasks. MapReduce breaks a computation into small tasks that run in parallel on different machines, scaling easily to several cluster. The core research activities of the project consist mainly in extending the execution and programming model to make Byzantine fault-tolerant MapReduce applications.