A Taxonomy of Clock Synchronization Algorithms Emmanuelle Anceaume and Isabelle Puaut

نویسنده

  • Emmanuelle Anceaume
چکیده

Clock synchronization algorithms ensure that physically dispersed processors have a common knowledge of time. This paper proposes a taxonomy adapted to all published software fault-tolerant clock synchronization algorithms: deterministic and prob-abilistic, internal and external, and resilient from crash to Byzantine failures. We classify clock synchronization algorithms according to their internal structure and to three othogo-nal and independent basic building blocks. Our taxonomy will help the designer in choosing the most appropriate structure of algorithm and the best building blocks suited to his/her hardware architecture, failure model, quality of synchronized clocks and message cost induced. Moreover, our classiication uses a uniform notation that allows to compare existing clock synchronization algorithms with respect to their fault model, the building blocks they use, the properties they ensure and their cost in terms of message exchanges. R esum e : Les algorithmes de synchronisation d'horloges oorent une notion commune du temps a des processeurs n'ayant pas acc es a une horloge globale partag ee. Ce rapport propose une classiication adapt ee a tous les algorithmes de synchronisation d'horloges evoluant dans un environnement sujet aux d efaillances, qu'ils soient d eterministes ou probabilistes, et qu'ils assurent une synchronisation interne ou externe. Les algorithmes de synchronisa-tion d'horloges etudi es sont class es selon leur structure interne et selon trois blocs de base ind ependants. Cette classiication est conn cue pour guider le concepteur d'un algorithme de synchronisation d'horloges dans le choix de l'algorithme le mieux adapt e a son architecture mat erielle, le mod ele de d efaillances vis e, la qualit e de la synchronisation obtenue ainsi que le co^ ut r esultant en terme de nombre de messages echang es. Par ailleurs, la classiication propos ee utilise une notation uniforme, qui permet de comparer les algorithmes existants selon le mod ele de fautes qu'il supportent, les blocs de base qu'ils utilisent, les propri et es qu'ils assurent et le co^ ut associ e.

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

Performance Evaluation of Clock Synchronization Algorithms

Clock synchronization algorithms ensure that physically dispersed processors have a common knowledge of time. This report proposes a survey of software fault-tolerant clock synchronization algorithms: deterministic, probabilistic and statistical ; internal and external ; and resilient from crash to Byzantine failures. Our survey is based on a classiica-tion of clock synchronization algorithms (...

متن کامل

HADES: A Middleware Support for Distributed Safety-Critical Real-Time Applications

Most distributed safety-critical real-time systems designed in the past have been specialized to meet the particular requirements of the application domain to which they were targeted. This approach led to specific, inflexible, dedicated and non-reusable solutions, often based on specialized hardware. This report presents an overview of HADES which provides a set of flexible tools built on top ...

متن کامل

HADES : A Middleware Support for Distributed Safety-Critical Real-Time Applicationsy

Most distributed safety-critical real-time systems designed in the past have been specialized to meet the particular requirements of the application domain to which they were targeted. This approach led to speciic, innexi-ble, dedicated and non-reusable solutions, often based on specialized hardware. This paper presents an overview of hades, which provides a set of exible tools built on top of ...

متن کامل

A Flexible Run-time Support for Distributed Dependable Hard Real-time Applications

Typically, most distributed, dependable, real-time systems designed in the past can only meet the particular requirements of the application domain to which they were targeted. This approach led to specific, non-flexible, dedicated and non-reusable solutions, often based on specialized hardware. This paper presents an alternative approach where a flexible run-time support for distributed depend...

متن کامل

Clock Synchronization in the Byzantine-Recovery Failure Model

We consider the problem of synchronizing clocks in synchronous systems prone to transient and dynamic process failures, i.e., we consider systems where all processes may alternate correct and Byzantine behaviors. We propose a clock synchronization algorithm based on periodical resynchronizations which is based on the assumption that no more than f < n/3 processes (with n the number of processor...

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 1997