DPCP (Discard Past Consider Present) -- A Novel Approach to Adaptive Fault Detection in Distributed Systems

  • Authors:

  • I. Sotoma

  • Affiliations:

  • -

  • Venue:
  • FTDCS '01 Proceedings of the 8th IEEE Workshop on Future Trends of Distributed Computing Systems
  • Year:
  • 2001

  • A survey of fault tolerant CORBA systems

    OTM'07 Proceedings of the 2007 OTM Confederated international conference on On the move to meaningful internet systems: CoopIS, DOA, ODBASE, GADA, and IS - Volume Part I

Quantified Score

Hi-index 0.00

Visualization

Abstract

Fault detection is a fundamental issue for faulttolerance in distributed systems.This paper presents theDPCP (Discard Past Consider Present) approach, thatdiscards the last elapsed times of fault detection messagesand considers only the current one.By this way, DPCPallows to perform a fast, accurated and scalable adaptivefault monitoring for asynchronous distributed systems.The scalability comes from the parameter MinimumTimeUnit,that controls the minimum frequency of thefault monitoring messages.The fastness and accuracy offault monitoring come from the changing of timeout andmonitoring interval values as soon as the systemworkload and the MinimumTime Unit allow.Some DPCPexperiments on ACE+TAO were made to observe DPCPbehavior on changing network workloads.