Artificial Intelligence for Cloud Computing Management

Recent years have seen a growing interest among users in the migration of their applications to the Cloud computing environments. However, due to high complexity, Cloud-based services often experience a large number of failures and security breaches, and consequently, impose numerous challenges on the dependability and resilience of users’ applications.

Unfortunately, current dependability and resilience solutions focus either on the infrastructure itself or on application analysis, but fail to consider the complex inter-dependencies between system components and application tasks.
This aspect is highly crucial especially when Cloud environments are used, as it is increasingly considered nowadays, in critical applications.

Besides, definition of application requirements, allocations of resources to application tasks, and optimization of global management parameters usually are based either on statistical approaches or on heuristics strategies typical of operating research. Computational intelligence may give additional opportunities and flexibility in specifying the requirements especially when they are defined by non-experts and in optimizing the resource allocation and the global management parameters.

This talk will discuss a user-centric, dependability- and resilience-driven framework that considers deploying and protecting users’ applications in the Cloud infrastructure so as to minimize their exposure to the vulnerabilities in the network, as well as offering fault tolerance and resilience as a service to the users who need to deploy their applications in the Cloud.

In this scenario, the talk analyzes the opportunities offered by computational intelligence to specify the characteristics and the requirements of these environments and support their management in the presence of many local optimization minima.