Information Technology Reference
In-Depth Information
The feasibility analysis performed in this approach, however, centres it on several
tools with the main purpose of providing numbers and graphical images that will ease
the decision of whether to tackle the migration or not. These tools and their purpose
are described next:
•
Code analysis:
The goal of such an analysis is twofold. On one hand, to represent
the coupling of methods, types, classes, namespaces or assemblies in varying sizes
according to common metrics like lines of code (LOC), McCabe's Cyclometric
Complexity, CBO (coupling between objects), RFC (response for class) [19]
RFC∞ [20] MPC (message passing coupling) [19] DAC (data abstraction coupling)
and DAC1 [21] ICP (information-flow-based coupling) [22] COF (coupling
factor), in order to obtain other useful views of code cohesiveness and complexity.
These views can be presented in a variety of ways such as a matrix, a dependency
graph or a tree. On the other hand, the second goal of this code analysis is to
discern the dependency of the legacy software in 3
rd
party COTS and/or with other
applications internal to the company.
•
High Level Modelling:
This activity has as main goal to obtain the understanding
of candidate functions or modules that might be exposed as services in an easy
manner. The best way to obtain this knowledge is by modelling the application with
UML in its different views, seizing the power of Reverse Engineering Techniques.
Also important at this stage is also to analyse the database schema, data and
transaction model in order to select the best database architecture (RDBMS,
NoSQL, a hybrid solution) and migration strategy towards the chosen one.
•
Effort estimation tool:
based on the desired target cloud platform, this tool
provides an estimation of the work (effort) that will be needed to transform it to
that target platform.
The main outcomes of this analysis are:
•
a set of metrics and indicators that show how complex the code is and thus how
much effort will be needed to transform the legacy application to a cloud oriented
environment. These metrics will be used in the Business Feasibility Phase to
extract the costs of the migration strategy,
•
a classification of legacy artefacts to be considered in a posterior Reverse
Engineering process, e.g.; source code, configuration files, data files,
documentation, existing models, etc.
•
based on the previous results, a taxonomy highlighting the main different types of
legacy artefacts according to their corresponding characteristic and properties.
•
effort calculation on the migration.
C. Business Feasibility Analysis
The goal of this business feasibility analysis is to provide guidelines that will aid the
management level take the decision whether to tackle the migration to SOA and/or
Cloud based on objective economic parameters. In order to do so, a cost-benefit
analysis is being developed (including ROI and Payback) that will cover the specific
