Data Centers are very dynamic environments hosting multiple types of server farms that all support key business applications. The design of the Data Center involves a variety of aspects related to how applications are architected, how they are deployed, and their network infrastructure.
A sound approach to design involves using a combination of architectural principles, such as scalability, flexibility, and high availability, as well as applying those principles to the requirements of the application environment. The result should be an architecture that meets the current needs but that is flexible enough to evolve to meet the needs of short- and long-term trends.
A solid foundation for Data Center design is based on a redundant, scalable, and flexible Layer 2 and Layer 3 infrastructure in which the behavior is both predictable and deterministic. The infrastructure also should accommodate service devices that perform key functions aimed at scaling or securing application environments. The deployment of service devices such as firewalls, load balancers, SSL offloaders, and caches requires careful planning.
The planning efforts must ensure that the desired behavior is achieved in the following areas: redundancy protocols between service devices, the exchange of connection and session information between stateful devices, the location of default gateway services, and the traffic path through the Data Center infrastructure from device to device.
Additional considerations require an architectural approach to deal with the application environment trends and the requirements that are imposed on the network infrastructure. Subsequent chapters in the book dig deeper into the specifics of Data Center and server farm designs.