Change Management: Data Center Management 101 Part V

Proper change management is not just a process, procedure, or committee but a commitment to excellence in your IT landscape. Those of you who already have some form of change management deployed inside your organization, my applause. Those who do not, I believe you will find this blog extremely useful in helping you understand more about the needs for robust change management in your environment.

The basics of change management ensure that prior to any change being implemented into a production environment, it undergoes a detailed assessment to determine potential impact and risks associated with that change. A simple example is the deployment of a new level of microcode on a switch or storage subsystem and the impact to this system (and the ones that use this system) if its upgrade goes awry, side-ways or simply fails and what measures are in place to account for these different occurrences (roll-back plans, alternative configurations, etc.).

Change management will ensure that all parties that could be affect are aware the change is occurring and have all the information they need to understand its impact to their systems prior to allowing the change to go forward. There are several categories of change management you should implement inside your IT Organization. These include: 

• Standard Change
• Normal day-to-day tasks. These tasks are frequently repeated, well documented and have a high probability of success (essentially guaranteed) with minimal degree of impact. It also assumes that the activity is being performed by a SME (Subject Matter Expert).
• Minor Change
• Minor impact to the infrastructure or application, again performed by the SME, where the risk to other systems is minimal or controlled.
• Significant Change
• Significant impact to the infrastructure or application, but will not impact the entire organization. Examples of these types of changes may include a technology refresh (storage system replacement) that impacts a significant number of applications in a specific location in the data center.
• Major Change
• Major impact to the infrastructure or application, likely to impact other areas of the organization. Examples of these changes include changes to the networking infrastructure or corporate email systems.
• Urgent Change
• Urgent impact, change needed to repair a portion of infrastructure and/or application that has failed or is not responding as expected due to unexpected load on the system.

To perform any of these changes should require varying level of approvals with more approvals needed as you move down the list. This will ensure all proper levels of management and business owners understand the full impact of any given change to the applications or business units for which they are responsible.

To determine who should do what, ITIL (Information Technology Infrastructure Library) includes some good examples but, for just get started, here are some individuals that you will need to involved in the process:

• Change Manager
• Change Advisory Board
• Management Board
• Change Advisory Board Emergency Committee 

Change management is a necessary evil in today's production data center environments. While it may add time and bureaucracy to everyday processes, the importance of keeping others informed about the changes that are occurring within a company's infrastructure and how these changes may potentially impact their applications or business unit cannot be understated. Change is a constant in today's data centers but only by implementing change management can companies ensure that these changes constantly occur without breaking existing processes or causing irreparable damage to corporate data or infrastructure equipment.