Showing papers by "Francesco Quaglia published in 2007"

A Lightweight Heuristic-based Mechanism for Collecting Committed Consistent Global States in Optimistic Simulation

Abstract: In this paper we study how to reuse checkpoints taken in an uncorrelated manner during the forward execution phase in an optimistic simulation system in order to construct global consistent snapshots which are also committed (i.e. the logical time they refer to is lower than the current GVT value). This is done by introducing a heuristic-based mechanism relying on update operations applied to local committed checkpoints of the involved logical processes so to eliminate mutual dependencies among the final achieved state values. The mechanism is lightweight since it does not require any form of (distributed) coordination to determine which are the checkpoint update operations to be performed. At the same time it is likely to reduce the amount of checkpoint update operations required to realign the consistent global state exactly to the current GVT value, taken as the reference time for the snapshot. Our proposal can support, in a performance effective manner, termination detection schemes based on global predicates evaluated on a committed and consistent global snapshot, which represent an alternative as relevant as classical termination check only relying on the current GVT value. Another application concerns interactive simulation environments, where (aggregate) output information about committed and consistent snapshots needs to be frequently provided, hence requiring lightweight mechanisms for the construction of the snapshots.

PELCR: Parallel environment for optimal lambda-calculus reduction

Abstract: In this article we present the implementation of an environment supporting Levy's optimal reduction for the λ-calculus on parallel (or distributed) computing systems. In a similar approach to Lamping's, we base our work on a graph reduction technique, known as directed virtual reduction, which is actually a restriction of Danos-Regnier virtual reduction.The environment, which we refer to as PELCR (parallel environment for optimal lambda-calculus reduction), relies on a strategy for directed virtual reduction, namely half combustion. While developing PELCR we adopted both a message aggregation technique, allowing reduction of the communication overhead, and a fair policy for distributing dynamically originated load among processors.We also present an experimental study demonstrating the ability of PELCR to definitely exploit the parallelism intrinsic to λ-terms while performing the reduction. We show how PELCR allows achieving up to 70--80p of the ideal speedup on last generation multiprocessor computing systems. As a last note, the software modules have been developed with the C language and using a standard interface for message passing, that is, MPI, thus making PELCR itself a highly portable software package.

Transparent Risk-free Synchronization in the High-Level-Architecture Interoperability Standard

Abstract: The high-level-architecture (HLA) is an IEEE standard for the interoperability and integration of (autonomous) simulation packages and applications (termed federates in the HLA context). It is based on a middleware-level component referred to as run-time-infrastructure (RTI) offering a set of interoperability services to the overlying simulation software. Time-management is the suite of services allowing synchronized execution among the federates, which, according to the HLA specification, covers pure conservative and pure optimistic synchronization schemes. In this paper we provide the design and implementation of a software layer, we refer to as risk-free-speculator (RFS), which supports an optimistic oriented intermediate approach to synchronization embedding the aggressiveness property of optimistic systems, but discarding risk. This is done in a totally transparent manner to the overlying applications, and does not even require any modification of the underlying RTI. The effectiveness of RFS has been tested against simulated demonstration exercises using the Joint Semi-Automated Forces (JSAF) simulation program.