Vinsamlegast notið þetta auðkenni þegar þið vitnið til verksins eða tengið í það: http://hdl.handle.net/1946/10205
Rapid scalability is important in cloud computing in order to serve growing communities and optimize hardware costs. This scalability can be hard to achieve, especially in software with static architecture. Changing software architecture of running systems on multiple devices over the Internet is a hard and delicate process as updating live software can cause faults and failures while software systems are being restarted. Taking the study of software architecture to the dynamics of the cloud computing can be beneficial in this case and increase cloud computing possibilities.
The Architectural Scripting Language (ASL) is a language for expressing the dynamic aspect of run-time and deployment-time software architecture. In the following thesis, ASL is taken to cloud computing which enables dynamic software architecture changes to meet the dynamics of a computing infrastructure. We present Cloud ASL, which is an external domain-specific language which enables architectural operations and architectural scripting in cloud computing environments. Cloud ASL is modeled and tested by the creation of a distributed cloud computing ray tracing system which was built to utilize Cloud ASL for its distributed and cloud computing mechanism.
Cloud ASL is a framework which enables modelling dynamic aspects of runtime software architecture with architectural operations in cloud computing and suitable to use for creating a scalable and modifiable cloud computing software.