Study on large-scale embedded databases evolution

Abstract

Dynamic software update research results mostly in the update the application code and memory of the state. With more and more application, to use embedded database to store, dynamic update solutions will have to support changes to embedded database format. The first step to support dynamic update the database format is to understand these embedded pattern change-so far, pattern evolution studies have focused on large enterprise database. In this paper, we put forward a kind of automatic extraction method from regular application embedded format, for example, written in C and C++ , automatic calculation mode changes how to apply the evolution. To show our method, we use a long-term pattern evolution research of the four popular open source program use embedded database and drab, Vienna. Our research accumulated some pattern across 18 evolution model, this paper reveals the change and frequency in embedded database mode change different enterprise database, formed the previous research object. This platform can be used to implement long term mass evolutionary study embedded schema potentially dynamic updates and schema for evolution researchers. © 2013 Springer-Verlag.