=Paper=
{{Paper
|id=Vol-1871/paper4
|storemode=property
|title=An Innovative Workspace for The Cherenkov Telescope Array
|pdfUrl=https://ceur-ws.org/Vol-1871/paper4.pdf
|volume=Vol-1871
|authors=Alessandro Costa,Eva Sciacca,Ugo Becciani,Piero Massimino,Simone Riggi,David Sanchez,Fabio Vitello
|dblpUrl=https://dblp.org/rec/conf/iwsg/CostaSBMRSV16
}}
==An Innovative Workspace for The Cherenkov Telescope Array==
https://ceur-ws.org/Vol-1871/paper4.pdf
8th International Workshop on Science Gateways (IWSG 2016), 8-10 June 2016
An Innovative Workspace for The Cherenkov
Telescope Array
Alessandro Costa ∗ , Eva Sciacca ∗ , Ugo Becciani ∗ , Piero Massimino ∗ ,
Simone Riggi ∗ , David Sanchez † and Fabio Vitello ∗
∗ INAF-Osservatorio Astrofisico di Catania, Italy
† Laboratoire d’Annecy-le-Vieux de Physique des Particules, France
Email: alessandro.costa@oact.inaf.it
Abstract—The Cherenkov Telescope Array (CTA) is an ini- Centre, which provides and disseminates data and analysis
tiative to build the next generation, ground-based gamma-ray software to the science community at large, using common
observatories. We present a prototype workspace developed at astronomical standards and existing computing infrastructures.
INAF that aims at providing innovative solutions for the CTA
community. The workspace leverages open source technologies The total data volume to be managed by the CTA Science
providing web access to a set of tools widely used by the CTA Data Centre is of the order of 27 PB/year [2], when all data-
community. Two different user interaction models, connected to set versions and backup replicas are considered. All levels of
an authentication and authorization infrastructure, have been data (from the raw data to the high-level final products) will be
implemented in this workspace. The first one is a workflow
archived in a standardised way, to allow access and reprocess-
management system accessed via a science gateway (based on
the Liferay platform) and the second one is an interactive virtual ing. The “CTA science gateway” will provide access to data,
desktop environment. The integrated workflow system allows to support services, software and data center infrastructures. It is
run applications used in astronomy and physics researches into foreseen that individual scientists using the analysis software
distributed computing infrastructures (ranging from clusters to made available by CTA can conduct the high-level analysis
grids and clouds). The interactive desktop environment allows
of CTA data. The Gateway aims at supporting workflow
to use many software packages without any installation on local
desktops exploiting their native graphical user interfaces. The handling, virtualization of hardware, visualization as well as
science gateway and the interactive desktop environment are resource discovery, job execution, access to data collections,
connected to the authentication and authorization infrastructure and applications and tools for data analysis.
composed by a Shibboleth identity provider and a Grouper Access to the developed services within the “CTA science
authorization solution. The Grouper released attributes are
consumed by the science gateway to authorize the access to
gateway” and other CTA web resources will be based on
specific web resources and the role management mechanism in each users profile and category (e.g. unsigned user, guest
Liferay provides the attribute-role mapping. observer, advanced user, principal investigator, archive user,
Keywords—Workflow Systems; Science Gateways; Collabora- pipeline user, etc). For such a purpose the Authentication
tive Environments; Astrophysics; DCIs and Authorization Infrastructure (AAI) plays a key role in
the scientific process and will be widely discussed in this
I. I NTRODUCTION paper. The AAI is also a fundamental part of the workspace
The Cherenkov Telescope Array (CTA) project1 aims at developed by INAF and described in this work.
building a new observatory for very high-energy (VHE) The “CTA science gateway” is implemented as a set of
gamma rays [1]. CTA has ambitious science goals focused in complementary modules. Three of them are being developed
understanding the origin of cosmic rays and their role in the with different aims: the first one is developed by INAF and
Universe, the nature and variety of particle acceleration around presented in this paper, it provides a workflow management
black holes and in searching for the ultimate nature of matter system, it is powered by WS-PGRADE/gUSE2 [3], based on
and physics beyond the Standard Model. For reaching these Liferay platform 3 and an added value of this module is a web-
goals it is aimed to achieve full-sky coverage by deploying desktop environment (with a VNC-based User Interface); the
hundreds of telescopes at two sites in the southern and the second one integrates existing CTA applications in a specific
northern hemispheres. InSilicoLab platform [4] developed by Cyfronet; and the third
To guarantee the smooth running of the complex CTA obser- module, developed by the Observatoire de Paris, is compliant
vatory three main management elements have been identified: with the Virtual Observatory and it is based on the Django
(i) the Science Operation Centre, which is in charge of the platform.
organisation of observations, (ii) the Array Operation Centre, The INAF prototype workspace is composed by a science
which conducts the operation, monitors the telescopes and the gateway module and by the Authentication and Authorization
atmosphere, and provides all calibration and environmental Infrastructure. The science gateway, first introduced in [5],
data necessary for the analysis, and (iii) the Science Data
2 WS-PGRADE/gUSE web page: http://guse.hu
1 CTA project web page: https://portal.cta-observatory.org 3 Liferay web page: https://www.liferay.com
8th International Workshop on Science Gateways (IWSG 2016), 8-10 June 2016
allows the user to access a workflow management system nector) [8], Globus8 [9], gLite9 [10], UNICORE10 (Uniform
with a customizable graphical web user interface (see Section Interface to Computing Resources) [11], PBS11 as well as web
II-A) and a web-desktop environment (see Section III) which services or clouds [12].
allows the use of plenty of software packages for both on-line The web-desktop environment: Astronomical & Physics
and off-line analysis of astrophysical data also exploiting their Cloud Interactive Desktop (ACID), allows to use many soft-
native Graphical User Interface. The gateway has been further ware packages without any installation on local desktops
developed to enable the processing of the Fermi Workflow exploiting their native Graphical User Interface. Finally, a
Demonstrator (see Section II-B) and connected to the Authen- common menu bar has been added to allow the integration
tication and Authorization Infrastructure developed by INAF with the different “CTA science gateway” modules. The role
(see Section IV). Moreover the gateway has been combined of the common menu-bar is to i) standardize the layout of
and integrated with the other modules put in place by the each module using a shared Common cascading Style Sheet;
Observatoire de Paris and Cyfronet using a common menu-bar ii) prompt the user with a common top-menu that provides
and the shared authentication and authorization infrastructure. access to each module; iii) prompt the user with information
about her/his current session: such as username and log-out /
II. T HE INAF CTA S CIENCE G ATEWAY log-in facilities.
The overall architecture of the INAF Catania workspace is
The INAF CTA science gateway4 is aiming at providing depicted in Figure 1. The gateway is able to connect with a
a web instrument for high energy astrophysics. It leverages variety of DCIs thanks to the integrated workflow management
on open source technologies giving web access to a set of system (see Section II-A), the embedded ACID environment
tools and software widely used by the CTA community. An allows both on-line and off-line analysis through commonly
extended (though not exhaustive) list5 of tools provided by this used CTA tools (see Section III). It is connected to the AAI
technology embrace XANADU software package, GammaLib to open the access to the CTA community accordingly to each
& ctools, Fermi Science Tools, Aladin, IRAF. Each tool is user own role and/or access right.
available interactively via a dedicated web-desktop environ-
ment or through a workflow management system.
Fig. 1. INAF Catania Workspace Architecture.
The gateway is based on the Liferay platform. Liferay
is an enterprise-level framework, offering both an advanced
development infrastructure and a flexible content management
system. We used the Liferay community edition that is released
under an open source GNU LGPL license. This provides
a cutting-edge and inexpensive solution that best suits our
purposes; Liferay is moreover used by a wide community
of users. These aspects are an important added value for a
technology destined to follow the CTA consortium for its
lifetime. Liferay platform has a large set of configurations
implementing High Availability (HA) solutions. The resulting
Liferay system will be able to handle the expected number of
concurrent users and subsequent traffic, and will reduce single
points of failure resulting in a more robust system. Liferay
can also be configured to load balancing and clustering at A. Workflow Management System
the server level. Both user profile management and workspace Scientific workflow management systems [13] offer means
applications are provided by the Liferay platform and can be to compose and distribute steps needed to perform com-
easily improved and customized according to the CTA present putations for data analysis or simulations, whereas hiding
and future requirements. INAF CTA science gateway provides details about the complex infrastructures underneath [14], [15].
a workflow management system with a customizable graphical More importantly, workflow descriptions capture the process
web user interface [6] and a web-desktop environment. of scientific experimentation, which are useful to reproduce,
The integrated workflow system (based on gUSE/WS- reuse or re-purpose these processes [16].
PGRADE) seamlessly enables the execution of astronomical A plethora of mature workflow systems has evolved
and physics workflows (and jobs) on major platforms such that support diverse concepts and languages with different
as DIRAC (Distributed Infrastructure with Remote Agent strengths and focus on different modes of processing. Few
Control) INTERWARE6 [7], ARC7 (Advanced Resource Con- workflow systems deliver the power of diverse digital re-
sources and most of the web-based creation and editing tools
4 INAF CTA science gateway: http://cta-sg.oact.inaf.it/
5 List of Software and tools available in the INAF CTA science gateway: 8 GLOBUS web page: https://www.globus.org/
http://acid.oact.inaf.it/ACID/Included packages.html 9 gLite web page: http://cern.ch/glite
6 DIRAC INTERWARE web page: http://diracgrid.org/ 10 UNICORE web page https://www.unicore.eu/
7 ARC web page: http://www.nordugrid.org/arc/ 11 PBS (Portable Batch System) web page http://www.pbsworks.com/
2
8th International Workshop on Science Gateways (IWSG 2016), 8-10 June 2016
either require local software installations with inherent security • the instrument response functions (IRFS).
problems or offer incomplete functionalities. Therefore gUSE The output of this use case are a set of FITS files returning
has been selected mainly because of: i) its usability via the processed maps.
web-based user interfaces; ii) its availability, with respect On the science gateway two workflows have been imple-
to licensing terms and cost; iii) its anticipated long-term mented: BINNED and UNBINNED running on to the INAF
support, e.g. via an active open-source community; and iv) Astrophysical Observatory of Catania clusters: muoni-server-
its ability to deal efficiently with the scales of data, com- 02.oact.inaf.it, and acid.oact.inaf.it. The workflows have been
putation and concurrent use required [17]. gUSE enables designed to set the input datasets and the parameters to run
users convenient and easy access to distributed computing the process into the InputSet job so that only the entry job
infrastructures (clusters, DIRAC INTERWARE, clouds) by is configured and then the parameters are passed to the other
providing a general purpose, workflow-oriented web-based jobs automatically.
user interface WS-PGRADE consisting of web services for The separation of the Fermi processing into different jobs
the workflow management and accessing various distributed within the workflow allowed us to exploit the full paralleliza-
data storages. tion of the computations within the configured DCIs. Finally
the OutputFileSet job collects all the jobs output files (typically
B. Fermi Workflow Demonstrator
in the order of tens of MB in this kind of analyses) and send
A demonstrator has been implemented following a typical them also to the ownCloud14 server which is synchronized
Fermi analysis performed with the Fermi Science Tools12 . with the user account hosted into the ACID environment.
Fermi Science Tools and data analysis chain is used to simulate Currently a step-up authentication allows the access to the
CTA analysis. The tools and data are publicly available on- ownCloud server and the integration with the AAI is foreseen
line13 . Figure 2 gives the diagram of the test case. Two as a future work.
analysis chains are available and should give similar results.
The 2 chains (BINNED and UNBINNED) share few steps in Fig. 3. BINNED workflow developed within the INAF CTA science gateway.
common. These chains are standard analysis steps for CTA
analysis.
Fig. 2. Fermi analysis test case.
III. ACID
The Astronomical & Physics Cloud Interactive Desktop15
[18] allows to use many software packages without any
The Fermi analysis requires 3 input files: an event file in installation on local desktops. Through the ACID environment
FITS format, a spacecraft file in FITS format, and the sky the users are able to exploit the native Graphical User Interface
model in XML format. of the available applications. A long list of astronomical
The user has to provide also few parameters for the analysis: and physics software suites are already available in ACID
including among others: ctools & GammaLib, Fermi Science
• the position of the target : Right Ascension (RA in
Tools, Geant4 PGPlot. Moreover it uses ownCloud to easily
degrees) and Declination (Dec in degrees),
share data between the user device and the ACID server(s).
• the energy ranges in MeV (Minimum energy Emin and
ACID is exploited by the science gateway offering two
maximum energy Emax),
modes of usage (see Figure 4):
• the time range in MET (Mission Elapsed Time, start time
and stop time), • on-line analysis: to perform interactive analysis through
• the radius of the region of interest to use (ROI in degrees), the Virtual Network Computing (VNC) and the native
GUIs or shell environment of the CTA applications. It
12 Fermi Science Tools web page: http://fermi.gsfc.nasa.gov/ssc/data/
analysis/scitools/overview.html 14 Owncloud web site: http://www.owncloud.com
13 Fermi Data: http://fermi.gsfc.nasa.gov/ssc/data 15 ACID web site: http://acid.oact.inaf.it/ACID/Home page.html
3
8th International Workshop on Science Gateways (IWSG 2016), 8-10 June 2016
Fig. 4. ACID usage modes from the science gateway: on-line analysis and national research and education network (NREN) organisa-
off-line analysis. tions and the European Union. eduGAIN interconnects identity
federations around the world, simplifying access to content,
services and resources for the global research and education
community. It enables the trustworthy exchange of information
related to identity, authentication and authorisation by coordi-
nating elements of the federation technical infrastructure and
providing a policy framework that controls this information
exchange.
We see eduGAIN as the best approach to achieve a CTA
consortium-wide authentication Infrastructure since the major-
ity of the consortium members already belong to eduGAIN.
INAF is therefore running an AAI (Authentication and
Authorization Infrastructure) composed by a Shibboleth17 CTA
Identity Provider and a Grouper18 authorization solution fully
compatible with the eduGAIN standards. The INAF CTA
science gateway module is able to handle users authenticated
Fig. 5. Geographical distribution of CTA consortium members.
by multiple federated Identity Provides, this is done by a Shib-
boleth Discovery Service19 . The CTA Identity Provider acts as
a cross-border/cross-domain CTA access complementing the
eduGAIN identity federation and granting access to each CTA
consortium user even in the case he is not (yet) member of
any national federation.
Federations participating in eduGAIN adhere to a common
lightweight technical and policy infrastructure and post their
local federation policies so that others can learn about their
registration practices and other relevant details. Each national
federation already publishes a trust registry in the form of a
metadata file. Each federation sends its registry to eduGAIN,
except the entries that a member organization does not want
to be included. eduGAIN combines all the national registries
and republishes them in one large file (see Fig. 6). A national
federation imports the eduGAIN consolidated international
registry, merges it with our local entries and publishes them
for your use.
does not require any installation from the user since the
browser loads and launches ACID as a Java Applet.
Fig. 6. eduGAIN structure.
• off-line analysis: through workflow submissions to ACID
employed as Distributed Computing Infrastructure (e.g.
as a node of a cluster). In this case a workflow job can
be configured to exploit the large number of command
line-based software packages available in the ACID en-
vironment.
IV. AUTHENTICATION AND AUTHORIZATION
I NFRASTRUCTURE
The CTA consortium is an experimental scientific collabora-
tion, it consists of over 1200 members working in 32 countries
from 200, mostly academic, institutes. The geographical loca-
tion of consortium members, as shown in Figure 5, leads to the
need of a pervasive Federated Identity Management network.
A. Authentication
17 Shibboleth web site: https://shibboleth.net/
eduGAIN16 is a service developed within the GANT net- 18 Grouper web site: http://www.internet2.edu/products-services/
work Project which is a major collaboration between European trust-identity-middleware/grouper/
19 Shibboleth Discovery Service web site: https://wiki.shibboleth.net/
16 eduGAIN web site:http://services.geant.net/edugain/Pages/Home.aspx confluence/display/SHIB2/DiscoveryService
4
8th International Workshop on Science Gateways (IWSG 2016), 8-10 June 2016
1) Discovery Service: In the CTA INAF science gateway Fig. 7. Authentication & Authorization Infrastructure.
the user can be authenticated by multiple IdPs (in principle
each IdP of the eduGAIN inter-federation); this is currently
done by using the Shibboleth Discovery Service. The Dis-
covery Service is the process by which a service provider
identifies the appropriate Identity Provider. In the case shown
in this paper, the discovery service watches the federation
metadata providing the user with a list identity providers. It
also provides a “Search-as-you-type” selection: an effective
search that guides the user in creating and reformulating
her/his selection.
B. Authorization
On top of the eduGAIN federated authentication infrastruc-
ture INAF Catania is providing a CTA authorization solution
based on Grouper. Grouper is an authorization solution that
keeps the membership affiliation consistent across multiple
applications allowing to create and manage groups. Groups
are used within each CTA application (e.g. a science gateway,
the archive user interface or the project management portal)
to track an individual role, or to determine which users are
authorized to access the resources. If groups are managed requests passing them to the gateway using AJP23 (Apache
separately in each application, keeping the membership list JServ Protocol).
consistent across these services becomes very difficult. The Shibboleth Service Provider has been configured to
Grouper provides a way to define a group once and use include the attribute prefix as ”AJP ”, otherwise user attributes
that group across multiple applications managing it at a single from Shibboleth could not be accessible in the gateway.
point. The single point of control implies that, once a person It was set the AJP communication with the backend and
is added or removed from a group, the group-related privileges configured Shibboleth to be ”activated” for the whole gateway
are automatically updated in all of the collaborative appli- and required a Shibboleth session at the login.
cations. The current Grouper prototype proposed by INAF We have so far identified two roles:
Catania is designed to manage and release the isMemberOf at-
• Advanced WF User: Creates new workflows with the
tribute and the eduPersonEntitlement attribute. These attributes
science gateway WorkFlow management User Interface
are consumed by the CTA INAF science gateway and are
• WF User: Defines the workflow parameters, launches
used to authorize the access to specific web resources. These
workflow processes and check results
attributes are released using standard SAML assertions. Figure
7 shows the architectural diagram of the Grouper-based AAI The attribute “isMemberOf” is mapped within the gateway
prototype. control panel as shown in the Figure 8 and the role permissions
are set at portlet level in order to allow/deny access to
specific gateway functionalities. In this case the “Advanced
C. Connection with the Science Gateway
WF User” role will have full access to the whole workflow
To connect the INAF CTA science gateway with the pro- management system functionalities (e.g. workflow design and
posed AAI an open-source Shibboleth Plugin available on implementation) while the “WF User” will be enabled only
GitHub20 has been employed and configured within the gate- to import pre-defined workflows developed by the advanced
way. The plugin provides an interface on the Liferay Control workflow users and customize them to run.
Panel, it allows for mapping of attributes and, thank to the
role mapping feature of Liferay it has been exploited for both V. R ELATED W ORK
authentication and authorization. Related prototyping works in the context of the “CTA sci-
The INAF CTA science gateway has been protected with ence gateway” activities comprise two other modules namely:
Shibboleth by running Apache HTTP21 server in front of the Cyfronet InSilicoLab module and the Data Distiller module
the Tomcat22 servlet container. It has been configured to run developed by the Observatoire de Paris. The InSilicoLab
on a private address and the Apache server intercepts all gateway supports Monte Carlo simulations performed on
distributed computing infrastructures (grids) thanks to the
20 Liferay Shibboleth Plugin: https://github.com/ivan-novakov/ integration with the DIRAC middleware. The Data Distiller
liferay-shibboleth-plugin
21 Apache HTTP web site: https://httpd.apache.org 23 AJP documentation: https://tomcat.apache.org/tomcat-7.0-doc/config/ajp.
22 Tomcat web page: http://tomcat.apache.org html
5
8th International Workshop on Science Gateways (IWSG 2016), 8-10 June 2016
Fig. 8. Shibboleth configuration within the INAF CTA science gateway. the possibilities of ACID for on-line interactive analysis and
off-line processing. We presented the Fermi Workflow Demon-
strator used as a test-bench for the typical CTA analyses.
We highlighted the possibilities for exploiting the full job
parallelization of the workflow management system and the
connection with the ACID environment for cloud storage of
output results. Finally we detailed the implementation and
usage within the science gateway of the authentication and
authorization infrastructure which guarantee the access to CTA
users tuned according to his/her own role and/or access rights.
Wide adopted standards (such as SAML 2.0 and Shib-
boleth 2.0) and open-source technologies (such as WS-
PGRADE/gUSE and Grouper) have been adopted within the
proposed workspace. This aims at enlarging the developer
community and improving the sustainability of the workspace
during the whole CTA lifetime. The proposed solution pro-
vides an highly flexible ecosystem in order to tailor a product
suitable to the present and future requirements of the CTA
community.
is implemented using the Python based Django framework 24 . The next steps within this work are foreseen to be focused
This prototype allows to search, retrieve and analyse high level on the integration of the proposed workspace with the other
CTA data products using the Virtual Observatory standards 25 . modules and services of CTA. In particular the consortium is
The aforementioned modules will be kept separated and an focusing on solutions to provide messaging protocols between
homogeneous user experience will be guaranteed using the the different modules. Moreover we will give support for the
same style-sheet. This will offer a common graphical inter- integration with the developed AAI to the other modules.
face including a menu-bar to navigate between the different
modules. An AAI assures to maintain the same user profiles ACKNOWLEDGMENT
and sessions; a message-passing paradigm will be employed The authors would like to thank colleagues from the
to share user data between the modules. CTA DATA Management group, in particular Nadine Ney-
Regarding the AAI for CTA, apart from the one discussed roud, Bruno Khelifi from LAPP (FR); Tomasz Szepieniec,
in this paper, a UNITY 26 (UNIfied identiTY) management Joanna Kocot, Hubert Siejkowski from Cyfronet (PL); a
prototype is under testing. It supports multiple authentication part of this work was developed within SCI-BUS (FP7-
protocols to allow integration with various consumers/clients INFRASTRUCTURES-2011 contract 283481) project.
and has the ability to outsource credentials management to a
3rd party service. The Grouper-based AAI has been preferred R EFERENCES
by the authors instead of the UNITY one because of its
reliability: it is deployed at many Universities and other [1] B. Acharya, M. Actis, T. Aghajani, G. Agnetta, J. Aguilar, F. Aharonian,
M. Ajello, A. Akhperjanian, M. Alcubierre, J. Aleksić et al., “Introduc-
organizations on 4 continents. Furthermore it provides features ing the cta concept,” Astroparticle Physics, vol. 43, pp. 3–18, 2013.
not currently available in UNITY such as the connection with [2] G. Lamanna, L. Antonelli, J. Contreras, J. Knödlseder, K. Ko-
the System for Cross-domain Identity Management27 (SCIM) sack, N. Neyroud, A. Aboudan, L. Arrabito, C. Barbier, D. Bastieri
et al., “Cherenkov telescope array data management,” arXiv preprint
or VOOT28 (an extensible protocol for dynamic exchange of arXiv:1509.01012, 2015.
group and authorization data), bulk user import/export and user [3] P. Kacsuk, Z. Farkas, M. Kozlovszky, G. Hermann, A. Balasko,
data (memberships) expiration. K. Karoczkai, and I. Marton, “WS-PGRADE/gUSE Generic DCI gate-
way framework for a large variety of user communities,” Journal of Grid
VI. C ONCLUSIONS AND F UTURE W ORKS Computing, vol. 10, no. 4, pp. 601–630, 2012.
[4] M. Hanasz, K. Kowalik, A. Gawryszczak, and D. Wóltański, “High per-
In this paper, we have introduced a workspace tailored to the formance astrophysical fluid simulations using insilicolab framework,”
requirements of the CTA community. The workspace consists in eScience on Distributed Computing Infrastructure. Springer, 2014,
pp. 293–304.
of a science gateway module based on the Liferay framework [5] A. Costa, P. Massimino, M. Bandieramonte, U. Becciani, M. Krokos,
endowed of a workflow management system and embedding a C. Pistagna, S. Riggi, E. Sciacca, and F. Vitello, “An innovative science
web-desktop environment (ACID). The workspace provides an gateway for the cherenkov telescope array,” Journal of Grid Computing,
vol. 13, no. 4, pp. 547–559, 2015.
authentication and authorization infrastructure. We described [6] A. Balasko, Z. Farkas, and P. Kacsuk, “Building science gateways by
24 Django web site: https://www.djangoproject.com utilizing the generic WS-PGRADE/gUSE workflow system,” Computer
25 IVOA web site: http://www.ivoa.net
Science, vol. 14, no. 2, pp. 307–325, 2013.
[7] A. Casajus, K. Ciba, V. Fernandez, R. Graciani, V. Hamar, V. Mendez,
26 UNITY web site: http://unity-idm.eu
S. Poss, M. Sapunov, F. Stagni, A. Tsaregorodtsev et al., “Status of the
27 SCIM web site: http://www.simplecloud.info
dirac project,” in Journal of Physics: Conference Series, vol. 396, no. 3.
28 VOOT: http://openvoot.org IOP Publishing, 2012, p. 032107.
6
8th International Workshop on Science Gateways (IWSG 2016), 8-10 June 2016
[8] M. Ellert, M. Grønager, A. Konstantinov, B. Kónya, J. Linde-
mann, I. Livenson, J. L. Nielsen, M. Niinimäki, O. Smirnova, and
A. Wäänänen, “Advanced resource connector middleware for lightweight
computational grids,” Future Generation computer systems, vol. 23,
no. 2, pp. 219–240, 2007.
[9] I. Foster and C. Kesselman, “The globus toolkit,” The grid: blueprint
for a new computing infrastructure, pp. 259–278, 1999.
[10] E. Laure, F. Hemmer, F. Prelz, S. Beco, S. Fisher, M. Livny, L. Guy,
M. Barroso, P. Buncic, P. Z. Kunszt et al., “Middleware for the next
generation grid infrastructure,” Computing in High Energy Physics and
Nuclear Physics (CHEP 2004), 2004.
[11] V. Huber, “Unicore: A grid computing environment for distributed and
parallel computing,” in Parallel Computing Technologies. Springer,
2001, pp. 258–265.
[12] M. Kozlovszky, K. Karóczkai, I. Márton, P. Kacsuk, and T. Gottdank,
“Dci bridge: Executing ws-pgrade workflows in distributed computing
infrastructures,” in Science Gateways for Distributed Computing Infras-
tructures. Springer, 2014, pp. 51–67.
[13] J. Liu, E. Pacitti, P. Valduriez, and M. Mattoso, “A survey of data-
intensive scientific workflow management,” Journal of Grid Computing,
vol. 13, no. 4, pp. 457–493, 2015.
[14] A. Belloum, M. Inda, D. Vasunin, V. Korkhov, Z. Zhao, H. Rauwerda,
T. Breit, M. Bubak, and L. Hertzberger, “Collaborative e-science ex-
periments and scientific workflows,” Internet Computing, IEEE, vol. 15,
no. 4, pp. 39–47, 2011.
[15] S. Olabarriaga, G. Pierantoni, G. Taffoni, E. Sciacca, M. Jaghoori,
V. Korkhov, G. Castelli, C. Vuerli, U. Becciani, E. Carley et al.,
“Scientific workflow management–for whom?” in e-Science (e-Science),
2014 IEEE 10th International Conference on, vol. 1. IEEE, 2014, pp.
298–305.
[16] U. Becciani, E. Sciacca, A. Costa, P. Massimino, C. Pistagna, S. Riggi,
F. Vitello, C. Petta, M. Bandieramonte, and M. Krokos, “Science
gateway technologies for the astrophysics community,” Concurrency and
Computation: Practice and Experience, 2014.
[17] P. Kacsuk, Science Gateways for Distributed Computing Infrastructures.
Springer, 2014.
[18] P. Massimino, A. Costa, U. Becciani, F. Vitello, and E. Sciacca, “Acid:
an interactive desktop for cta science gateway,” in Science Gateways
(IWSG), 2014 6th International Workshop on. IEEE, 2014, pp. 55–60.
7