Abstract
Cloud computing offers flexible IT solutions that suites the growingly dynamic organizational demands. Different types of cloud deployment models offer distinctive solutions based on the security requirements of the organization. Hybrid clouds are considered the prominent deployment model that offers the extended flexibility of public cloud resources and the security of private clouds. This paper aims at modelling and evaluating three firewall deployment models for hybrid clouds: no firewall, regular firewalls and special firewalls that block web traffic. Experimental results illustrated that having a regular firewall would be sufficient for small hybrid clouds (<= 150 nodes) to give high point-to-point utilization without compromising the response time. However, for larger networks (up to 300 nodes), a firewall that blocks web traffic is more efficient.