Hybrid cloud computingmonitoring software architecture

摘要

Cloud computing provides scalable computing resources on demand.Monitoring cloud computingresources is important so that resources can be dynamically allocated, migrated, or shut downto meet users' requirements. Challenges in cloud computing monitoring systems include detectingpatterns that might lead to failure of the cloud system, detecting malfunctioning problemswithin the cloud platform after they occur, and issues related to the fact that existingmonitoringsolutions are tightly coupled to specific cloud platforms, while there is an emerging need formonitoringsolutions that are independent of underlying subsystems. To address these challenges, wedesigned a hybrid cloud monitoring software architecture that can work as an add-on layer ontop of existing cloud computing platforms. The proposed architecture is designed based on facadesoftware design pattern and utilizes complex event processing concept inwhich data from variousprimitive metrics streams is processed to detect previously defined patterns. Prototype applicationshave been developed to demonstrate the architecture's usability. Performance tests wereapplied to prototype applications. Computation times required for the operation of the proposedarchitecture were negligible.