Cloud Computing Architecture - Deployment Models

The video starts by continuing the discussion on cloud architecture, focusing on different service and deployment models in cloud computing, including public, private, hybrid, and community clouds.

Public clouds are available to the general public, accessible over the internet, and are managed by third-party providers. Examples include Google App Engine, Microsoft Azure, and Amazon EC2. Key features include hidden workload locations, risks from multi-tenancy, network dependency, limited visibility/control over data, elasticity, and low upfront costs. However, they come with restrictive default service level agreements (SLAs).

Private clouds are provisioned for exclusive use by a single organization. They can be owned, managed, and operated by the organization or outsourced. Private clouds offer greater control and security, with examples like Eucalyptus and OpenStack. They can be on-site or outsourced, each with its own set of characteristics regarding network dependency, IT skill requirements, hidden workload locations, multi-tenancy risks, data import/export limitations, strong security, and significant upfront costs.

An outsourced private cloud maintains control under the organization's policies but is managed off-premise by a third party. It involves two security perimeters and relies on strong, protected communication channels. Benefits include extensive resource availability, though network dependency and multi-tenancy risks persist. Upfront costs are typically modest to significant.

Community clouds serve a specific community of organizations with shared concerns, such as missions, security requirements, or policy compliance. They can operate in public or private modes, on or off-premises. Access policies can be complex, often using role-based or attribute-based access control. Community clouds offer potentially strong security from external threats due to shared policies but may have high variable upfront costs.

Hybrid clouds combine two or more distinct cloud infrastructures (private, community, or public). This model allows organizations to keep sensitive data on private clouds while using public clouds for less critical workloads or for scaling. Portability of data and applications, and interoperability between different cloud types, are crucial considerations. Hybrid clouds can be extremely complex to manage due to variations in performance, reliability, and security across constituent clouds, and the dynamic nature of cloud components joining or leaving.

The choice of deployment model depends entirely on an organization's requirements, including security needs, budget, and application types. Small organizations might prefer public clouds, while those with critical data might opt for private or a hybrid approach. The decision involves classifying applications and data into categories to determine which cloud model is most suitable for each, presenting a significant management challenge.