When All the Observers of a Distributed Computation Do Agree

نویسندگان

  • Michel Raynal
  • Eddy Fromentin
چکیده

Architectures parallèles, bases de données, réseaux et systèmes distribués When all the observers of a distributed computation do agree Abstract: A consistent observation of a distributed computation can be seen as a sequence of states and events that might have been produced by executing this computation on a monoprocessor. So a distributed execution generally accepts a lot of consistent observations. This paper concentrates on what all these observations have in common. An abstraction called inevitable global state is deened. A necessary and suucient condition characterizing such states is given and a monitor-based algorithm that detects them is also presented. Possible uses of such states are sketched. Quand tous les observateurs d'une ex ecution r epartie sont d'accord R esum e : On peut consid erer qu'une observation coh erente d'une ex ecution r epar-tie se compose d'une s equence d' etats et d' ev enements qui aurait pu ^ etre produite en d eroulant cette ex ecution r epartie sur un mono-processeur. Une telle ex ecution r e-partie accepte donc g en eralement un grand nombre d'observations coh erentes. Nous etudions ici une abstraction que nous appellerons etat global in evitable qui est un point commun a toutes les observations. Nous donnons une condition n ecessaire et suusante qui caract erise cette propri et e d'in evitabilit e etant donn e un etat global, pr esentons un algorithme reposant sur l'emploi d'un contr^ oleur centralis e qui d etecte de tels etats globaux et en r esumons quelques utilisations possibles.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

When All the Observers of a Distributed Computation Do Agree Eddy Fromentin and Michel Raynal

A consistent observation of a distributed computation can be seen as a sequence of states and events that might have been produced by executing this computation on a monoprocessor. So a distributed execution generally accepts a lot of consistent observations. This paper concentrates on what all these observations have in common. An abstraction called inevitable global state is deened. A necessa...

متن کامل

Computation and Spacetime Structure

We investigate the relationship between computation and spacetime structure, focussing on the role of closed timelike curves (CTCs) in promoting computational speedup. We note first that CTC traversal can be interpreted in two distinct ways, depending on ones understanding of spacetime. Focussing on one interpretation leads us to develop a toy universe in which no CTC can be traversed more than...

متن کامل

Self-Triggered and Event-Triggered Set-Valued Observers

This paper addresses the problem of reducing the required network load and computational power for the implementation of Set-Valued Observers (SVOs) in Networked Control System (NCS). Eventand self-triggered strategies for NCS, modeled as discrete-time Linear Parameter-Varying (LPV) systems, are studied by showing how the triggering condition can be selected. The methodology provided can be app...

متن کامل

تعیین شاخص‌های ارزیابی سیستم‌های اطلاعات بهداشتی و ارائه الگو

Introduction: Health information systems have designed in order to manage health information fluency for improving healthcare quality. It is necessary to conduct continues evaluations to do epidemiologic researches and manage health information systems, enhance quality and decrease costs. Unfortunately there isn’t a framework for that foeus on the measurement methods and indicators in our count...

متن کامل

Signature-Free Communication and Agreement in the Presence of Byzantine Processes (Tutorial)

Communication and agreement are fundamental abstractions in any distributed system. (If the computing entities do not need to communicate or agree in one way or another, the system is not a distributed system!) This tutorial was devoted to the design of such abstractions built on top of signature-free asynchronous distributed systems prone to Byzantine process failures. It is made up of three p...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 1994