Information Technology Reference
In-Depth Information
Summary
Migrating an existing application to the cloud is a complex and multi-
dimensional problem requiring in many cases adapting the application in
significant ways. Taking a look in particular into the database layer of the
application, this involves dealing with differences in the granularity of inter-
actions, refactoring of the application to cope with remote data sources, and
addressing data confidentiality concerns. In this chapter we introduce an
application migration methodology that incorporates these aspects, and a
decision support, application refactoring and data migration tool which sup-
ports application developers in realizing this methodology. We evaluate the
proposed methodology and enabling tool using a case study conducted in
the context of an e-science project.
5.1 Introduction
e-Science is an active field of research striving to enable faster scientific dis-
covery and groundbreaking research in different scientific domains by means
of information technology (IT). It is considered a new paradigm for science
and is referred to as the fourth paradigm (Hey et al., 2009) or data-intensive
science ; it unifies theory, experiments, and simulation for data exploration
for the purpose of scientific discovery. Existing literature shows that myriad
available software systems, like Kepler, Triana, Taverna, Pegasus, and so on,
support only some of the experiment life cycle phases and are applicable
only for specific scientific domains (Taylor et al., 2006).
Due to its interdisciplinary nature, e-science exhibits a high degree of
complexity, mainly due to the technical challenges and interoperability defi-
ciencies of the existing software, the large amounts of data produced and
consumed by the computational tools and systems, and the computational
intensity and distributed characteristics of the IT environment observed in
scientific computing. One major issue in current research is the integration of
existing software and tools, across domains and organizational structures,
for enabling the collaborative modeling of more complex scientific experi-
ments and their execution. The most prominent approach for integrating
software systems for the purpose of performing scientific experiments is
workflow technology. Workflows are defined in terms of control flow among
tasks comprising an experiment and the data exchanged among them
(i.e., data flow). Moreover, the tasks in a workflow stand for a concrete unit of
work that can be implemented by a computational, configuration, or visual-
ization tool or by human users.
 
Search WWH ::




Custom Search