Abstract
Cloud computing is one of the most discussed areas in computer science in the last years. Although there is an extensive amount of research covering this field, the field of bio-inspired cloud computing is underinvestigated when compared to the general research area. This study tries to find answers on how a biomorphic model can be implemented in the cloud in order to achieve adaptive cloud behaviour. The process of cellular differentiation where cells transform from one type to another, is chosen to be the foundation model for a developed technical model. We define analogies to the cloud where stem cells are blank servers and web servers are cells with a specific function. With a combination of configuration management, version control and cloud deployment systems, an imitation of this biological process is applied in the cloud. The use of automated cloud scaling as a case of adaptive behaviour is the main goal of the research. Two different approaches have been developed for mapping the biological model to the cloud. The first approach consists of a prototype where the signal detection and node activation is being triggered by using the concept of random generated timers. The second approach is based on the concept of random seeds which are used to coordinate the transformation procedure. The project results were able to adapt the cloud based on current needs, with each prototype having its advantages and disadvantages over the other.