Deployment Framework Best Practices for Red Hat Enterprise Linux on HP ProLiant

4
The scale and the longevity of the business service being provided are important in the development
of a deployment framework. For these criteria, there are often three differing strategic approaches for
deployment processes:
• Divergence - for environments comprised of quick cycles and short life times of the resulting
service. These processes focus heavily on efficient, automated installation processes with little
concern for long-term configuration management and maintenance tasks, instead favoring a
reinstall of the then latest approved operating environment. For this reason, divergence-
based methodologies tend to rely upon native or vendor provided tools with as little
integration as possible. Speed and agility from a starting point is the driving concern for
selection of deployment framework components.
• Congruence - where the business service being offered is comprised of a very small set, but
has longevity. Systems all go through a prescribed set of steps, typically starting at a known
image-like snapshot. Even errant changes are captured and replayed along with the rework
steps toward the current configuration, and all systems are directed to stay in
synchronization. Since this process flow is all orchestrated, a combination of tool-based and
solution oriented offerings are typically employed. Consistency through detection of new
steps, usually via automation, is the driving concern for deployment frameworks utilizing this
methodology.
• Convergence - for environments that focus on consistency of business services across
various operating platform. The starting point, even operating platform, is of little concern
and you can treat as a black box. Combinations of tool-based and solution-oriented offerings
are typically included, and due to the business logic associated with the service
configuration, custom or point tools are likely in use for the integration. The key tenet of
convergence is to manage the business service configuration to behave as desired result.
Therefore, it is important to map an environment's tendencies and assess future changes before
attempting to develop a deployment framework. Considering the number of operating platforms, the
integration culture, and the scaling tendencies will minimize any change management issues.
Recognition of the desired attributes also filters out choices that are not compatible with current and
future directions. Any quadrant, and mix of continuums, can have successful deployment framework
instances.
Framework scope
In order to develop an effective deployment strategy, the next aspects under consideration are what
key artifacts of the traditional IT infrastructure are included. It is important for your deployment
strategy to determine which are in scope, as it really is the boundary layers between components that
you are trying to manage with your deployment framework. Even if starting with a minimal set, you
must anticipate how to manage more components over time, so that the deployment framework scales
appropriately for your environment.
Figure 2 shows the portions of the overall IT model that you can visualize as a stacked layer
representation.