S.Co.P.E. is one of the four projects funded by the Italian Government in order to provide Southern Italy with a distributed computing infrastructure for fundamental science. Beside being aimed at building the infrastructure, S.Co.P.E. is also actively pursuing research in several areas among which astrophysics and observational cosmology. We shortly summarize the most significant results obtained in the first two years of the project and related to the development of middleware and Data Mining tools for the Virtual Observatory.
Deep Dive into Astrophysics in S.Co.P.E.
S.Co.P.E. is one of the four projects funded by the Italian Government in order to provide Southern Italy with a distributed computing infrastructure for fundamental science. Beside being aimed at building the infrastructure, S.Co.P.E. is also actively pursuing research in several areas among which astrophysics and observational cosmology. We shortly summarize the most significant results obtained in the first two years of the project and related to the development of middleware and Data Mining tools for the Virtual Observatory.
arXiv:0807.0967v1 [astro-ph] 7 Jul 2008
Mem. S.A.It. Vol. 00, 0
c⃝SAIt 2008
Memorie della
Astrophysi s
in
S.Co.P
.E.
M. Brescia1, S. Cavuoti2, G. d’Angelo2, R. D’Abrusco2, C. Donalek3
N. Deniskina2, O. Laurino2, Giuseppe Longo2,1,4
1 INAF - Osservatorio Astronomico Capodimonte, via Moiariello 16, 80131, Napoli, Italy
2 Department of Physical Sciences - University Federico II, Naples, Italy
3 Department of Astronomy, California Institute of Technology, Pasadena, CA, USA
4 INFN - Napoli Unit, via Cintia 9, 80126, Napoli, Italy
Abstract. S.Co.P.E. is one of the four projects funded by the Italian Government in order to
provide Southern Italy with a distributed computing infrastructure for fundamental science.
Beside being aimed at building the infrastructure, S.Co.P.E. is also actively pursuing re-
search in several areas among which astrophysics and observational cosmology. We shortly
summarize the most significant results obtained in the first two years of the project and re-
lated to the development of middleware and Data Mining tools for the Virtual Observatory.
Key words. distributed computing, cosmology
1. Introduction
S.Co.P.E. is a general purpose GRID infras-
tructure of the University Federico II in Naples
funded through the Italian National Plan
(PON) by the Italian Government to support
both fundamental research and small/medium
size companies. The infrastructure has been
conceived as a metropolitan GRID, embedding
different (and in some cases pre-existing) and
heterogeneous computing centers each with its
specific vocation: high energy physics, astro-
physics, bioinformatics, chemistry and mate-
rial sciences, electric engineering, social sci-
ences. Its intrinsically multi-disciplinary na-
ture renders the S.Co.P.E. an ideal test bed for
innovative middleware solutions and for inter-
operable tools and applications finely tuned on
the needs of a distributed computing environ-
Send offprint requests to: G. Longo, e-mail:
longo@na.infn.it
ment. In what follows we shall shortly out-
line the main activities in the fields of astro-
physics and observational cosmology and, in
particular, we shall focus on: i) the ongoing ef-
forts aimed at integrating the S.Co.P.E. GRID
(hereafter SG) with the international Virtual
Observatory (Sect.2), and ii) the implementa-
tion in the SG of the data mining (DM) VO-
Neural package (Sect.3) which is developed
in the framework of a collaboration with the
Dept. of Astronomy at Caltech. In Sect. 4 we
shortly outline a template scientific application
and, finally, in Sect. 5, we outline some future
developments.
2. The VOb and the GRID
The Virtual Observatory (VOb) is an in-
ternational effort coordinated through the
International Virtual
Observatory Alliance
(IVOA; URL.1) aimed at: i) federating and
M. Brescia & et al.: astrophysics in S.Co.P.E.
1
making interoperable all astronomical data
archives produced by both ground based and
space borne instruments; ii) deploying a new
generation of science applications or tools
which use VOb protocols for exploratory data
analysis and for the extraction of knowl-
edge from massive data sets. The VOb is in-
herently distributed: data collections remain
with their providers and are accessed through
standard interfaces. The access to the data
takes place through a registry which con-
tains information about data sets, archives,
catalogs, surveys, and computational services
that can be accessed through VOb interfaces
(Hanish & De Young 2008). While the federa-
tion and fusion of heterogeneous data archives
and the implementation of flexible data reduc-
tion and data analysis tools have been widely
addressed and, at least in their fundamen-
tal aspects, solved, the possibility to access
large distributed computing facilities to per-
form computing intensive tasks has not yet
been satisfactorily answered.
One problem to be solved is the conflict ex-
isting between the VOb and the GRID secu-
rity procedures: most users of a specific Virtual
Organization (VO) do not possess the personal
certificates which are requested to access the
GRID or, even when they do have a personal
certificate, the computing GRID which they
need does not recognize their own certification
authority.
In the framework of the VONeural project
(Sect. 3) and in order to make our Data Mining
(DM) tools accessible to the wider commu-
nity, we implemented and tested a general pur-
pose interface between the UK-ASTROGRID
(hereafter AG; URL.4) and the SG.
2.1. GRID-Launcher v.1.0
The UK based ASTROGRID is one of the most
robust astronomical Virtual Organizations so
far implemented and represents a good ground
for testing innovative solutions. The main
problem we had to face was the fact that most
users which are recognized by the AG User
Authentication Service do not possess a per-
sonal GRID certificate and cannot therefore
access distributed computing resources. This
Fig. 1. Grid Launcher v.1.0 workflows for input
and output. UI: user interface; RB: resource broker;
SE: storage eleme
…(Full text truncated)…
This content is AI-processed based on ArXiv data.