Semantic and Algorithmic Recognition Support to Porting Software Applications to Cloud - Trustworthy Eternal Systems via Evolving, Software Data and Knowledge - page 7

4 Design of the Architecture and Ongoing
Implementation of the Porting Support Procedure
The architecture implementing the methodology described in the previous sec-
tion, is illustrated in Fig. 2 with the workflow and interactions among the com-
ponents, while Fig. 3 illustrates the workflow for the API annotation process.
Fig. 2. Workflow of the porting procedure
The architecture is composed of the following four modules.
The ALCOR (ALgorithmic COncept Recognizer) module [10] recognizes algo-
rithmic concepts in the code, producing the Abstract Program Representation.
The API Ontology Builder extracts the graph by parsing source code and from
the graph representing the code produces the API grounding ontology which
represent the base where to ground the semantic annotations. This ontology
enables the annotation of the API elements in a simple way, by adding relation
between grounding elements and high level abstraction concept. Additionally
the API Ontology Builder produces the Candidate API ontology graph applying
graph transformation patterns defined for the specific programming language or
model.
The Schema Matcher module [14] accepts two graph based representations
and performs the matching between the two graphs, by applying several algo-
rithm including structural-based algorithms and syntactical ones.
The Annotator allows the user to semantically annotate the target API with
concepts from the functional ontology.