Designing supercontinuum spectra using Grid technology A. Ferrando* a , C. Milián b , N. González a,c , G. Moltó b , P. Loza d , M. Arevalillo-Herráez e , M. Zacarés f , I. Torres-Gómez c , V. Hernández b a Departament d’Òptica, Universitat de València, Dr Moliner 50, 46100 Burjassot (València), Spain; b Instituto de Instrumentación para Imagen Molecular, Universidad Politécnica de Valencia, Camino de Vera S/N, 46022 Valencia, Spain; c Centro de Investigaciones en Óptica, A.C., Guanajuato (León), México; d ICFO-Institut de Ciències Fotòniques, Mediterranean Technology Park, 08860 Castelldefels (Barcelona), Spain; e Departament d’Informàtica, Universitat de València, Avda Vicent Andrés Estellés S/N, 46100 Burjassot (València), Spain; f Departamento de Ciencias Experimentales y Matemáticas, Universidad Católica de Valencia, C/ Guillem de Castro 94, 46003, Valencia, Spain. ABSTRACT We present a new computational scheme to design supercontinuum spectra “à la carte” by means of Genetic Algorithms. Due to the potentially large amount of computations required by this strategy, the deployment of these heuristic algorithms is performed using distributed computing in the form of a Grid platform. The optimization procedure is automated within the Grid platform and permits escalation to large computational Grids. Some examples of designed supercontinua are given and potential applications for the design of future photonic devices are briefly described. Keywords: Supercontinuum, optimization, genetic-algorithms, Grid technology 1. INTRODUCTION The generation of optical homogeneous spectra covering a range from the near infrared (IR) to the ultra violet (UV) is known as Supercontinuum (SC) generation. This phenomenon is usually observed in photonic crystal fibers and it has been a subject of intense study in the field of temporal nonlinear optics for the last decade. Nowadays there is a deep understanding of the underlying physics associated to this process [1, 2], opening a scope for engineering problem proposals. Since its discovery lots of applications have been found in optical metrology, pulse compression and white light sources generation, among many others. Our interest focuses on nonlinear microscopy of biological samples, where it is particularly important to excite certain transition of a given protein or any other biological molecule. This technique needs very specific wavelengths which, in general, can not be efficiently provided by any standard source. Hence, the possibility of maximizing the spectral output of a fiber in an arbitrarily chosen spectral window is extremely useful. This kind of applications can be mathematically stated as an optimization problem that consists of determining the set of configuration parameters which yields the best results. Unfortunately, the evaluation of each combination of parameters is computationally expensive and the exploration of the entire space is an unfeasible option. In this context, the use of meta-heuristics approaches, such as genetic algorithms, are an alternative to approximate a global optimum. Nevertheless, the amount of required computations can become too large for standard computational resources. We have chosen Grid computing technologies [3] as the suitable platform to implement our numerical codes. Grid technology is one of the most promising infrastructures in computing science. One of its main and most attractive feature is to provide user friendly access to large scale distributed computational facilities, overcoming the traditional drawbacks of standard super computing. This feature includes standard protocols for sharing both computational power and data storage capacity among geographically distant resources. In addition, the very nature of the Grid concept incorporates the scalability of the applications running on it. Therefore, they are an ideal infrastructure for the execution of the high throughput problem that lies beneath the optimization via Genetic Algorithms. *albert.ferrando@uv.es; www.interttech.upv.es Invited Paper 2nd Workshop on Specialty Optical Fibers and Their Applications (WSOF-2), edited by Juan Hernández-Cordero, Ismael Torres-Gómez, Alexis Méndez, Proc. of SPIE Vol. 7839, 78390W · © 2010 SPIE · CCC code: 0277-786X/10/$18 · doi: 10.1117/12.867203 Proc. of SPIE Vol. 7839 78390W-1