Description: Cloud computing (CC) (Buyya et al., 2009), a natural extension of grid computing is a new concept of distributed computing, which has emerged recently. It assumes that customers do not own or rent any computational infrastructure but they simply use available resources and pay for what they use. The potential platforms - solutions of CC concept will base on two classes of computational resources:

1. private or public computing clusters,
2. desktop workstations present in companies or universities, which are known to be idle around 50% of the time.

To transform the ideas of CC into specific platforms with wide acceptance, a number of issues must be solved. Despite classical problems in this domain the implementation of CC concepts needs confidentiality in software and user data, integrity on programs and data before, during and after run on CC platform and trust management to monitor resources and guarantee a given quality of service and security to users (Jafar et al., 2009), (Mahjoub et al., 2010). These issues focused little academic interests until know in greed computing, and lack of solutions in this field may explain industry’s reluctance to adopt grid computing despite the initial interest.
The purpose of the research topic is to develop a generic model and build environment of secured and trusted distributed computing in CC environment characterized by fault-tolerance phenomena of computational infrastructure. The work will be focused on secure and robust load balancing and scheduling algorithms which fulfill not only optimization criteria, like e.g. a minimal computational time but also provide confidentiality of data and programs, authentication of users and resources, resistance to attacks and malicious behavior of other actors of CC and take into account trust management criteria (Święcicka et al., 2006), (Szaban and Seredyński, 2010), (Ostaszewski et al., 2007). To build computational models and design secure and effective algorithms solving NP-hard nature problems a set of nature inspired paradigms will be used. This will include cellular automata, artificial immune networks, evolutionary algorithms, etc. Also computational frameworks and tools like game theory and multi-objective optimization will be useful.