Peer-to-Peer Algorithms for a Distributed Chemical Machine - DUMAS - Dépôt Universitaire de Mémoires Après Soutenance Accéder directement au contenu
Mémoires Année : 2011

Peer-to-Peer Algorithms for a Distributed Chemical Machine

Résumé

The present document reports the work done during the first four months of an internship hosted by the MYRIADS team under supervision by Cédric Tedeschi. The aim of the internship is to study various algorithms that can help to create a distributed runtime for the Chemical Programming Paradigm. The Chemical Programming model is an intrinsically parallel, non deterministic programming paradigm offering abstraction and expressiveness. It is based on a metaphor between rewriting rules and chemical reactions: bits of data are rewritten to other bits just as molecules are replaced by other molecules. During the internship, a model of Distributed Chemical Machine has been developed. This model includes a modified reaction process and a distribution scheme. A simulator has been implemented so as to test some features of the Distributed Chemical Machine. The report gives context to the subject, explaining why a distributed runtime for the Chemical Programming Paradigm is highly relevant in the Internet of Services we experience daily. Chemical Programming is then described from both an intuitive and formal point of view. A section thoroughly details the specification of our Distributed Chemical Machine. Finally, the simulator for this model is presented along with results it helped obtain. The conclusion gives insight in the future developments we expect about distribution of the Chemical Programming Paradigm.
Fichier principal
Vignette du fichier
Proust.pdf (411.82 Ko) Télécharger le fichier
Loading...

Dates et versions

dumas-00636788 , version 1 (28-10-2011)

Identifiants

  • HAL Id : dumas-00636788 , version 1

Citer

Raphael Proust. Peer-to-Peer Algorithms for a Distributed Chemical Machine. Operating Systems [cs.OS]. 2011. ⟨dumas-00636788⟩
296 Consultations
592 Téléchargements

Partager

Gmail Facebook X LinkedIn More