Our approach to data center relocations isolates and compartmentalizes data sets, minimizes known variables, and arranges potential back-out points. In a recent post, we shared three must-do tasks before any data center relocation. They were:
- Know your current data center inventory
- Understand the data center relocation’s impact on your business’ bottom line
- Know your risks
While these broad tasks are crucial, when performing a hot data center relocation – moving existing equipment to a new location or establishing new equipment in a new location without downtime – there are many more variables at play than a cold data center relocation. The following 18 tasks will help create a smoother transition, resulting in fewer opportunities for data center downtime:
- Interview key personnel regarding data-center relocation requirements. Different stakeholders will have different sets of needs, some similar and some different. Don’t assume you know what they are.
- Identify data-center relocation business priorities, needs, and constraints. While everyone will tell you that their requirements are the most important, it’s up to you and your IT team to determine priorities, along with what’s feasible and what’s not.
- Manage the exit plan with the source facility. Your current data center should be able to help you facilitate your exit plan and accommodate for any unique needs your data center move entails.
- Manage the installation plan with the target facility. Like your exit plan, your installation plan will need to involve the target facility and its team. They can help make your needs a reality.
- Startup servers and applications of new system information (e.g., IP addresses, name etc.). Be sure to keep track of all new system information so there are no surprises on switch-over.
- Assist in verification of properly working system and applications. You need to make sure that new systems are operating as planned prior to switch-over.
- Track and document amount of data to be replicated. Remember—data multiplies exponentially, especially with big data and the Internet of Things (IoT).
- Track and document changes made to data being replicated. This will ensure that all changes are accounted for and nothing is missed. It also ensures your data integrity.
- Recommend the amount of bandwidth needed to maintain a replication of their applications. Don’t underestimate bandwidth requirements at this phase; overestimate if need be.
- Ensure equivalent amount of storage available at new site. Calculate the amount of storage you’ll need at your new site to accommodate for the current data storage.
- Establish IP connection between the two sites.
- Document the external facing IP addresses at old and new sites. This step, while perhaps obvious, can help with troubleshooting and ensure a seamless transition.
- Install and implement replication (VVR, SNDR, HORC). This will help with any data backup and recovery needs that arise.
- Begin replication; continue asynchronous replication until fully replicated.
- Prepare DNS propagation within the WWW / internal.
- Switch to synchronous replication to ensure transaction integrity.
- Complete the switch over.
- Manage the entire project, including the formal project plan and coordination of all parties. Take responsibility and be sure to manage the process from start to finish.
Data center relocations and moves are, by nature, a stressful task to undergo, but with the right processes in place, they can (and do) go smoothly.