Separability to help parallel simulation of distributed computations

Authors:
Philippe Mauran;Gérard Padiou;Philippe Quéinnec
Affiliations:
Institut de Recherche en Informatique de Toulouse, UMR, CNRS, Université de Toulouse, ENSEEIHT, Toulouse cedex 7;Institut de Recherche en Informatique de Toulouse, UMR, CNRS, Université de Toulouse, ENSEEIHT, Toulouse cedex 7;Institut de Recherche en Informatique de Toulouse, UMR, CNRS, Université de Toulouse, ENSEEIHT, Toulouse cedex 7
Venue:
OPODIS'07 Proceedings of the 11th international conference on Principles of distributed systems
Year:
2007

Citing 9
Cited 0

Virtual time

ACM Transactions on Programming Languages and Systems (TOPLAS)
Complexity of network synchronization

Journal of the ACM (JACM)
Efficient algorithms for distributed snapshots and global virtual time approximation

Journal of Parallel and Distributed Computing - Special issue on parallel and discrete event simulation
Exploiting temporal uncertainty in parallel and distributed simulations

PADS '99 Proceedings of the thirteenth workshop on Parallel and distributed simulation
Time, clocks, and the ordering of events in a distributed system

Communications of the ACM
Elements of distributed computing

Elements of distributed computing
Principles of Distributed Systems

Principles of Distributed Systems
Seven-O'Clock: A New Distributed GVT Algorithm Using Network Atomic Operations

Proceedings of the 19th Workshop on Principles of Advanced and Distributed Simulation
Local-lag and timewarp: providing consistency for replicated continuous applications

IEEE Transactions on Multimedia

Quantified Score

Hi-index	0.00

Visualization

Abstract

We consider the parallel simulation of distributed systems based upon the notion of separability. We define a model of distributed systems that integrates a limited set of temporal properties, from which we deduce specific parallel simulation strategies avoiding rollback and dynamic scheduling which are often necessary to obtain an optimal rate of parallelism in parallel and/or distributed simulations. In particular, we present strategies that enable to compute the simulation schedule in advance, or even statically. This approach appears to be relevant for a large class of distributed computations, inasmuch as it relies upon a limited set of temporal properties of the simulated systems.