α-Register

  • David Bonnin
  • Corentin Travers
Conference paper

DOI: 10.1007/978-3-319-03850-6_5

Part of the Lecture Notes in Computer Science book series (LNCS, volume 8304)
Cite this paper as:
Bonnin D., Travers C. (2013) α-Register. In: Baldoni R., Nisse N., van Steen M. (eds) Principles of Distributed Systems. OPODIS 2013. Lecture Notes in Computer Science, vol 8304. Springer, Cham

Abstract

It is well known that in an asynchronous message-passing system, one can emulate an atomic register providing that more than half of the processes are non-faulty. By contrast, when a majority of the processes may fail, simulating atomic register is not possible. This paper investigates weak variants of atomic registers that can be simulated tolerating a majority of processes failures. Specifically, the paper introduces a new class of registers, called α-register and shows how to emulate them.

For atomic registers, a read operation returns the last written value when there is no concurrent write operations. α-registers generalize atomic registers in the following sense: In any interval I, at most α values written before I are returned by the read operations in I. A simulation of an α-register tolerating f failures in a n-processes system is presented for α = 2M − 1, where M =  max (1,2f − n + 2). The simulation is optimal up to a constant multiplicative factor: the paper establishes that α-registers cannot be simulated tolerating f failures if α ≤ M.

Keywords

Message passing fault-tolerance shared-memory simulation 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer International Publishing Switzerland 2013

Authors and Affiliations

  • David Bonnin
    • 1
  • Corentin Travers
    • 1
  1. 1.LaBRI, University Bordeaux 1France

Personalised recommendations