Petros Manousis, Panos Vassiliadis, George Papastefanatos.

Impact Analysis and Policy-Conforming Rewriting of Evolving Data-Intensive Ecosystems

Abstract

Data-intensive ecosystems are conglomerations of data repositories surrounded by applications that depend on them for their operation. In this paper, we address the problem of performing what-if analysis for the evolution of the database part of a data-intensive ecosystem, to identify what other parts of an ecosystem are affected by a potential change in the database schema, and how will the ecosystem look like once the change has been performed, while, at the same time, retaining the ability to regulate the flow of events. We model the ecosystem as a graph, uniformly covering relations, views, and queries as nodes and their internal structure and interdependencies as the edges of the graph. We provide a simple language to annotate the modules of the graph with policies for their response to evolutionary events to regulate the flow of events and their impact by (i) vetoing (b components, depending on policy conflicts, and (iii) rewrite the modules to adapt to the change. We theoretically prove the coverage of the language, as well as the termination, consistency, and confluence of our algorithms and experimentally verify our methods effectiveness and efficiency.

Input

The "projects": WCS, CS, S, Drupal

The input file of the events is located in the OTHER directory of the "projects".

Output

Csv files containing the results:

• S with mixture developer policy.
• S with mixture database administrator policy.
• CS with mixture developer policy.
• CS with mixture database administrator policy.
• WCS with mixture developer policy.
• WCS with mixture database administrator policy.
• Drupal evolution with default schema.
• Drupal evolution with modified schema.

Source code

The source code of Hecataeus used in this paper is here.

Hecataeus project

More information about Hecataeus project is here.