Evolution pattern for Service Evolution in Clouds
Zhe Wang, Xiaodong Liu, Kevin Chalmers, Guojian Cheng · Research Output (Edinburgh Napier University) · 2012
The proposed research will focus ondeveloping a novel approach to solve Software ServiceEvolution problems in Computing Clouds. Theapproach will support dynamic evolution of thesoftware service in clouds via a set of discoveredevolution patterns. An initial survey informed us thatsuch an approach does not exist yet and is in urgentneed. Pattern will be categorised into five groups inorder to solve different kinds of evolution problems,such as SaaS dependency dealing, SaaS structuretransformation, SaaS evolvability enhancement, SaaSsystem dynamic enhancement and SaaS functionenhancement. It contains dynamic process and staticdetails. Evolution mechanism will be used for driventhe evolution pattern into execution; both mechanismand pattern are corresponded with each other. Afterthe pattern design is finished, the whole approach willhave these contributions: 1) the synergistic patterndrivenframework for cloud service evolution; 2) arepository of evolution patters for cloud services(mainly for SaaS); 3) the formal specification of theevolution patterns; 4) the mechanism to apply thepatterns to cloud service evolution at run time: drivingthe evolution, monitoring the process, and verifying thecorrectness; 5) the resultant prototype tool. Bothdynamic and static description of the evolution patternhas been given in the paper.