Migrating e-Science Applications to the Cloud: Methodology and Evaluation - Cloud Computing with e-Science Applications - page 96

consider the non-Salesforce-specific steps for our proposed methodology.
As discussed extensively in Section 5.4, Laszewski and Nauduri also pro-
posed a vendor-specific methodology for the migration to Oracle products
and services by providing a detailed methodology, guidelines, and recom-
mendations focusing on relational databases (Laszewski and Nauduri, 2011).
We base our proposal on their methodology, by abstracting from it and
adapting and extending it.
Apart from the vendor-specific migration methodologies and guidelines,
there are also proposals independent from a specific cloud provider. Reddy
and Kumar proposed a methodology for data migration that consists of the
following phases: design, extraction, cleansing, import, and verification.
Moreover, they categorized data migration into storage migration, database
migration, application migration, business process migration, and digital
data retention (Reddy and Kumar, 2011). In our proposal, we focus on the
storage and database migration as we address the database layer. Morris
specifies four golden rules of data migration with the conclusion that the
IT staff does not often know about the semantics of the data to be migrated,
which causes a lot of overhead effort (Morris, 2012). With our proposal of
a step-by-step methodology, we provide detailed guidance and recom-
mendations on both data migration and required application refactoring to
minimize this overhead. Tran et al. adapted the function point method to
estimate the costs of cloud migration projects and classified the applications
potentially migrated to the cloud (Tran et al., 2011). As our assumption is that
the decision to migrate to the cloud has already been taken, we do not con-
sider aspects such as costs. We abstract from the classification of applications
to define the cloud data migration scenarios and reuse distinctions, such as
complete or partial migration to refine a chosen migration scenario.
As we discuss the prototypical realization of a tool providing support
and guidelines while deciding for a concrete cloud data store or service, the
migration, and the refactoring of the application architecture accordingly, in
the following we also investigate the state of the art on decision support sys-
tems (DSSs) (Power, 2002) in the area of cloud computing. Khajeh-Hosseini
et al. introduced two tools that support the user when migrating an appli-
cation to infrastructure-as-a-service (IaaS) cloud services (Khajeh-Hosseini
et al., 2011). The first one enables the cost estimation based on a UML deploy-
ment model of the application in the cloud. The second tool helps to identify
advantages and potential risks with respect to the cloud migration. None of
these tools is publicly available. We do not consider the estimation of costs
or the identification of risks as our assumption is that the decision for migra-
tion to the cloud has already been taken. We consider aspects such as costs,
business resiliency, effort, and so on to be considered before following our
methodology and using the tool (Andrikopoulos et al., 2013). Menzel and
Ranjan developed CloudGenius, a DSS for the selection of an IaaS cloud
provider focusing on the migration of web servers to the cloud based on