Towards Federated Service Discovery and Identity Management in Collaborative Data and Compute Cloud Infrastructures

Abstract

This paper compares three multi-national research infrastructures, one that provides data services, one that provides compute services, and one that supports linguistics research. The aim is to jointly provide services to the user communities, and, perhaps eventually, seamlessly interoperate. To this end, we look at and compare how the infrastructures build their service federations (trust, service status, information systems), and how they manage users (identities, authentication, and authorisation).

Full text

J Grid Computing (2018) 16:663–681
https://doi.org/10.1007/s10723-018-9445-3
Towards Federated Service Discovery and Identity
Management in Collaborative Data and Compute Cloud
Infrastructures
Shiraz Memon ·Jensen Jens ·Elbers Willem ·
Helmut Neukirchen ·Matthias Book ·
Morris Riedel
Received: 18 May 2017 / Accepted: 11 June 2018 / Published online: 26 June 2018
© Springer Nature B.V. 2018
Abstract This paper compares three multi-national
research infrastructures, one that provides data ser-
vices, one that provides compute services, and one
that supports linguistics research. The aim is to jointly
provide services to the user communities, and, per-
haps eventually, seamlessly interoperate. To this end,
we look at and compare how the infrastructures build
their service federations (trust, service status, informa-
tion systems), and how they manage users (identities,
authentication, and authorisation).
H. Neukirchen ·M. Book
University of Iceland, Reykjavik, Iceland
H. Neukirchen
e-mail: helmut@hi.is
M. Book
e-mail: book@hi.is
S. Memon ()·M. Riedel
J¨ulich Supercomputing Centre, Forschungszentrum J¨ulich,
Leo-Brandt Straße, 52428 J¨ulich, Germany
e-mail: a.memon@fz-juelich.de
M. Riedel
e-mail: m.riedel@fz-juelich.de
J. Jens
STFC, Harwell Oxford Campus, Didcot, UK
e-mail: jens.jensen@stfc.ac.uk
E. Willem
CLARIN ERIC, Utrecht, Netherlands
e-mail: willem@clarin.eu
Keywords Distributed infrastructure ·Federated
identity management ·Service discovery ·
Standards ·Interoperation ·Cloud computing
1 Introduction
Distributed compute, data, and more recently, cloud
infrastructures have been successful in providing
resources to a wide variety of research communi-
ties. The e-Infrastructure Reflection Group identified
in 2004 the outline/vision of a distributed infrastruc-
ture comprised of fabric (disk, CPU, networks), and
a “middleware” layer connecting the infrastructure
across sites; user communities would then develop
and deploy their own applications on top of the e-
infrastructure [44]. Also the Foster/Kesselman vision
of grid computing [31], with computing available on
demand through standard interfaces, was hugely influ-
ential in the development and use of e-infrastructures,
leading for example to the middleware that is known
as Globus Toolkit [29] and more recent Globus cloud
services [30].
The established e-infrastructures have been very
successful, having provided resources to researchers
on a national or multinational scale in TeraGrid [36],
European National Grid Initiatives (NGIs), Extreme
Science and Engineering Discovery Environments
(XSEDEs) [52], or, in the case of the world-wide
Content courtesy of Springer Nature, terms of use apply. Rights reserved.