On applying coordinated atomic actions and dependable software architectures for developing complex systems

Modern concurrent and distributed applications are becoming increasingly complex; so, in order to provide fault tolerance, special structuring mechanisms are required to help reduce this complexity. Unfortunately, such structuring techniques are mostly introduced as design and implementation features, which complicates their employment. The approach we propose relies on introducing the appropriate software structuring together with associated fault tolerance measures at the earlier phases of software development and on supporting it with special software architectures and design patterns