A GRID solution for Gravitational Waves Signal Analysis from Coalescing Binaries: preliminary algorithms and tests F. Acernese 1,2, F. Barone 2,3, R. De Rosa 1,2, R. Esposito 2, P. Mastroserio 2, L. Milano 1,2, S. Pardi 1, K. Qipiani 2, G. Spadaccini 1,2 1 Dip. Scienze Fisiche, Univ. Napoli, Italy, I INFN, sez. Napoli, Italy, I Dip. Scienze Farmaceutiche, Univ. Salerno, Italy, I Motivation The procedure for the extraction of gravitational wave signals coming from coalescing binaries provided by the output signal of an interferometric antenna may require computing powers generally not available in a single computing centre or laboratory. In order to overcome this problem, one of the possible solutions is that of using the computing power available in different places as a single geographically distributed computing system. This possibility is now effective within the GRID environment, that allows distributing the required computing effort for specific data analysis procedure among different sites according to the available computing power. Within this environment we decided to develop a system prototype with application software for the first experimental tests of a geographically distributed computing system for the analysis of gravitational wave signal from coalescing binary systems. These tests have been performed within a small GRID environment based on three local nodes: the Bologna node, acting as Tier-1, and the Napoli and Roma nodes, both acting as Tier-2. The tests were aimed: 1- to get information both on the reliability and the performances of the physics application software developed for the coalescing binary analysis by the Napoli group; 2- to test the efficiency of integration special tools developed for GRID within this specific application.Motivation The procedure for the extraction of gravitational wave signals coming from coalescing binaries provided by the output signal of an interferometric antenna may require computing powers generally not available in a single computing centre or laboratory. In order to overcome this problem, one of the possible solutions is that of using the computing power available in different places as a single geographically distributed computing system. This possibility is now effective within the GRID environment, that allows distributing the required computing effort for specific data analysis procedure among different sites according to the available computing power. Within this environment we decided to develop a system prototype with application software for the first experimental tests of a geographically distributed computing system for the analysis of gravitational wave signal from coalescing binary systems. These tests have been performed within a small GRID environment based on three local nodes: the Bologna node, acting as Tier-1, and the Napoli and Roma nodes, both acting as Tier-2. The tests were aimed: 1- to get information both on the reliability and the performances of the physics application software developed for the coalescing binary analysis by the Napoli group; 2- to test the efficiency of integration special tools developed for GRID within this specific application. Conclusions  We have successfully verified that multiple jobs can be submitted and the output retrieved with small overhead time;  Computational grids seems very suitable to perform data analysis for coalescing binaries searches.Conclusions  We have successfully verified that multiple jobs can be submitted and the output retrieved with small overhead time;  Computational grids seems very suitable to perform data analysis for coalescing binaries searches.. GRID GRID is an infrastructure that enables the integrated, collaborative use of high-end computers, networks, databases, and scientific instruments owned and managed by multiple organisations. GRID applications often involve large amounts of data and/or computing and often require secure resource sharing across organisational boundaries, and are thus not easily handled by today's Internet and Web infrastructures. Within this context the Globus project is developing fundamental technologies needed to build computational grids.GRID Matched Filter Algorithm It is the best procedure in searching of known waveform embedded in noise background. Despite of its optimal character,it requires a high computational cost. In fact, this method is based on an exhaustive comparison between the signal and all the possible waveforms taken in account, called templates. If the number of templates increases, the quality of the signal identification increases but a very large number of operations is needed Matched Filter Algorithm It is the best procedure in searching of known waveform embedded in noise background. Despite of its optimal character,it requires a high computational cost. In fact, this method is based on an exhaustive comparison between the signal and all the possible waveforms taken in account, called templates. If the number of templates increases, the quality of the signal identification increases but a very large number of operations is needed Steps Steps of the procedure Step 1 The data were extracted from CNAF-Bologna Mass Storage System. The extraction process reads the VIRGO standard frame format, performs a simple resampling and publishes the selected data file on the Storage Element; Step 2 The search was performed dividing the template space in 200 subspace and submitting from Napoli User Interface a job for each template subspace. Each job reads the selected data file from the Storage Element (located at CNAF-Bologna) and runs on the Worker Nodes selected by Resource Broker in the VIRGO VO. Finally, the output data of each job were retrieved from Napoli User Interface Steps Steps of the procedure Step 1 The data were extracted from CNAF-Bologna Mass Storage System. The extraction process reads the VIRGO standard frame format, performs a simple resampling and publishes the selected data file on the Storage Element; Step 2 The search was performed dividing the template space in 200 subspace and submitting from Napoli User Interface a job for each template subspace. Each job reads the selected data file from the Storage Element (located at CNAF-Bologna) and runs on the Worker Nodes selected by Resource Broker in the VIRGO VO. Finally, the output data of each job were retrieved from Napoli User Interface Conditions of the test procedure Algorithm: standard matched filters Templates generated at PN order 2 with Taylor approximants Data Data simulated at 20 kHz Each data frame is 1 second long Total data length: 600 s Conditions raw data resampled at 2 kHz lower frequency: 60 Hz upper frequency: 1 kHz search space: 2 – 10 solar masses minimal match: 0.97 Number of templates: ~ Conditions of the test procedure Algorithm: standard matched filters Templates generated at PN order 2 with Taylor approximants Data Data simulated at 20 kHz Each data frame is 1 second long Total data length: 600 s Conditions raw data resampled at 2 kHz lower frequency: 60 Hz upper frequency: 1 kHz search space: 2 – 10 solar masses minimal match: 0.97 Number of templates: ~ Scheme of the computational GRID used during the tests Computing Element Worker Node 1 Worker Node 3 Storage Element Worker Node 2CNAF-Bologna Resource Broker Information Index Replica Catalogue Computing Element Worker Node 1 Worker Node 2 User Interface INFN Roma1 Computing Element Worker Node 1 Worker Node 2 User Interface INFN Napoli GARR Data Storage Element Map of the templates used in the test (tau space) Università degli Studi di Napoli “Federico II” Istituto Nazionale di Fisica Nucleare - Sezione di Napoli Università degli Studi di Salerno CHEP03 - March 24-28, La Jolla, California