Validating a Peer-to-Peer Evolutionary Algorithm

Abstract

Abstract. This paper proposes a simple experiment for validating a Peer-to-Peer Evolutionary Algorithm in a real computing infrastructure in order to verify that results meet those obtained by simulations. The validation method consists of conducting a well-characterized experiment in a large computer cluster of up to a number of processors equal to the population estimated by the simulator. We argue that the validation stage is usually missing in the design of large-scale distributed metaheuristics given the difficulty of harnessing a large number of computing resources. That way, most of the approaches in the literature focus on studying the model viability throughout a simulation-driven experimentation. However, simulations assume idealistic conditions that can influence the algorithmic performance and bias results when conducted in a real platform. Therefore, we aim at validating simulations by running a real version of the algorithm. Results show that the algorithmic performance is rather accurate to the predicted one whilst times-to-solutions can be drastically decreased when compared to the estimation of a sequential run.

Designing robust volunteer-based evolutionary algorithms

“…In order to reproduce such speedups in a real setting, we would require more demanding problems than those addressed in this paper. However, as we demonstrated in [Laredo et al(2012)], quasi-linear speedups are feasible in large-scale systems when tackling time-consuming fitness evaluation functions, i.e. an increasing ratio between computation and communication favors scalability.…”

The Seamless Peer and Cloud Evolution Framework

“…Finally, an abstract model for scalable P2P evolutionary computation, named EvAg (Evolvable Agent), was proposed in [17], and a thorough analysis of the scalability of the system with respect to the problem size was provided. This model was further investigated in [16]. However, in both studies, the authors focused mostly on the conceptual aspects of the distributed GA (and how to simulate those) rather than the actual implementation.…”