- Define for Success
- Identifying Automation Components
- Aligning Goals with Reality
- Planning Buy, Build, and Deployment
- Harris Kern's Enterprise Computing Institute
Planning Buy, Build, and Deployment
It's now time to move on to planning the design and deployment of the solution. While this phase obviously requires more tutelage than can be covered effectively in the scope of one article, a broad-stroke outline of some key points for consideration can help ensure the success of an automation project.
The design work for the automation of the technology architecture and the administrative processes should occur both separately and simultaneously. Doing this planning in tandem is important because effecting the automated provisioning of storage without considering change-management policies and protocols (for example) is a recipe for disaster.
Creating a design plan for the technical architecture basically comprises evaluating and choosing the system-management software to support the technical components identified for automation. This phase of the design planning should be completed with the participation of a project manager and will result in system-management software solutions that have implications for system, storage, and peripheral hardware; network hardware and software; and database applications.
The design process is fairly straightforward, but that doesn't mean that it's easy. As a starting point, however, it helps to use the automation criteria outlined in this piece to develop requests for proposals to be distributed to external vendors and staff. The responses will help you determine what's required to meet your automation goals and assess whether the solution should be purchased from an outside vendor, developed internally, or created by a combination of the two.
In this planning phase, give special consideration to the platform architecture requirements; features such as supportability and cost will have substantial impact on system-management software and the automation of operations. Additionally, multiple-platform requirements will be the most expensive to fulfill.
As you plan for technical architecture design and deployment, consider your budgetary constraints and decide whether outsourcing the product(s), implementation, and/or ongoing service is the most cost-effective way to proceed. Like most other aspects of data center automation, the outsourcing question is complex enough to fill an entire book (and has). A careful assessment of the technologies, processes, and financial considerations of your outsourcing project is in order when weighing this question.
Finally, designing the automation to interface with administrative processes is often overlooked. It involves more than the purchase of new tools. It requires taking an inventory of existing processes and refining them to support the new levels of automation. Without this work, the entire initiative can fail. In fact, its omission is the most common cause of failure for data center automation projects. So, at this phase, IT managers must be developing strategies to enact the following production process components:
Production assessmentprocess that identifies the operational requirements to implement and manage new and changing applications
Problem managementcentralized process to manage and resolve user network, application, and systems problems
Change managementprocess that coordinates all changes that affect the production environment
Asset managementprocess to query, discover, track, and store enterprise-computing resources, including hardware, operating systems, and applications
Disaster recoveryprocess to enable recovery in the event that a disaster should render mission-critical systems inoperable
Again, I cannot emphasize enough the importance of planning to interface heavily with these administrative processes. If you don't provide for the process redesign and training, your initiative will surely fail.
As the speed and reach of business grows, increasing numbers of companiesboth large and smallwill turn to data center automation to leverage new information opportunities, streamline processes, and cut costs. By understanding what automation means and entails, and planning appropriately for both the technical and administrative processes, IT managers can do much to ensure the validity and long-term success of their initiatives.