BG Distributed Simulation Algorithm - Archive ouverte HAL Accéder directement au contenu
Chapitre D'ouvrage Année : 2016

BG Distributed Simulation Algorithm

Résumé

How to effectively translate an algorithm from a distributed system model to another one? Distributed systems come in diverse settings, that are modeled by different assumptions 1) on the way processes communicate, e.g., using shared memory or messages , 2) on the fault model, 3) on synchrony assumptions, etc. Each of these parameters has a dramatic impact on the computing power of the model, and in practice, an algorithm or an impossibility result is usually tailored to a particular model and cannot be directly reused in another model. This wide variety of models has given rise to many different impossibility theorems and numerous algorithms for many of the possible combinations of parameters that characterize them. Thus, a crucial question is the following: are there bridges between some models, i.e., is it possible to transfer an impossibility result or an algorithm from one model to another? The Borowsky-Gafni simulation algorithm, or BG simulation, is one of the first step towards direct translations of algorithms or impossibility results from one model to another. The BG simulation considers distributed systems made of asynchronous processes that communicate using a shared memory array. In a nutshell, this simulation allows a set of t + 1 asynchronous sequential processes, where up to t of them can stop during their execution, to simulate any set of n ≥ t+1 processes executing an algorithm that is designed to tolerate to up to t fail-stop failures. The BG simulation has been used to prove solvability and unsolvability results for crash-prone asynchronous shared memory systems, paving the way for a more
Fichier principal
Vignette du fichier
bg-revised.pdf (240.41 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02087581 , version 1 (02-04-2019)

Identifiants

Citer

Matthieu Roy. BG Distributed Simulation Algorithm. Ming-Yang Kao. Encyclopedia of Algorithms, Springer New York, pp.199-203, 2016, 978-1-4939-2863-7. ⟨10.1007/978-1-4939-2864-4_611⟩. ⟨hal-02087581⟩
78 Consultations
332 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More