How to avoid common consolidation pitfalls.
With over a decade of experience executing data center consolidation projects, we are called on from time-to-time to salvage flailing projects, and we have seen a pattern of common, recurring consolidation errors. These errors routinely occur across all industry types.
Below are our findings organized into 8 common consolidation errors.
1. Inadequate or Absent Methodology.
A data center consolidation is not your everyday IT project. The nuances and complexities can wreak havoc on your operations. Your project is vulnerable to missed schedules, budget overruns, and operational failure.
Such a project demands a methodology specific to data center consolidations. Having one enables you to mitigate the potential errors and identify hidden risks specific to your data center consolidation project. It is both strategic and tactical.
A competent consolidation methodology uses the business priorities of the organization to drive the consolidation strategy. Think of it as an outside-in approach. It first seeks to understand the business processes and its’ associated applications then looks inward to develop the tactical plan using tried and true processes, tools, and procedures.
A mature and proven consolidation methodology helps to both guide the project and protect the enterprise. Without one, it’s like rolling the dice and hoping for the best.
2. Team Lacks Experience.
Most organizations do not have the expertise on staff to plan and execute a consolidation nor are they typically familiar with the complexities a data center consolidation presents. An inexperienced team can present a clear and immediate danger to your business operations. We know this first hand.
We often receive calls from organizations that tried to go it alone but later engaged us to rescue the failing project. A data center consolidation project demands experienced, subject matter experts. The costs and benefits associated with your data center consolidation project are directly related to how effectively the project team is staffed and managed.
3. Poor Communication.
A consolidation affects every person who is a part of your organization: business partners, clients, third party entities and end-users. Communication is essential. A comprehensive communications plan is key to avoiding not only operational outages but also the political complications that can choke a project. It is rarely managed well.
It is imperative that all stakeholders across the company have access to key information as it affects their daily work. Get your HR team involved early. It precludes the complications of rumors and encourages ownership from top to bottom.
Here are a few practical tips:
- Don’t overlook communicating hardware and network changes.
- Communicate with all departments during early planning stages to account for high processing times such as end of year, quarterly or monthly events.
- As you approach move day(s), prepare a final hour-by-hour go-live communication plan and distribute to affected stakeholders as required.
4. No Disaster Recovery or Backup Plan.
Managing risk is a critical part of every consolidation. Unforeseen event(s) and production problems can occur at a moment’s notice. Your disaster recovery and backup plan must be activation-ready.
The plan should include an in-depth assessment of key system interdependencies and critical business processes.
Risk factors to consider for your plan are:
- Loss of data
- Loss of power
- Geographic location of servers during consolidation
- Transportation of equipment
- Damage to equipment during transport
- Lack of resources to manage consolidation
- Acts of God (fire, flood, earthquake, etc.)
- Utility and technical concerns at new site (electrical, cooling, internet, cabling, racks, etc.)
We recommend meeting with your internal IT team, facilities department, network administrators, utility providers and other relevant stakeholders to ensure that all voices are heard regarding the “what if” scenarios.
5. Improve While You Move.
While most people believe consolidation is a great time to upgrade and refresh servers and business applications, we do not.
Yes, thoughts and plans must be given to an end state, but it’s risky enough to consolidate terabytes of data, servers and other infrastructure often dispersed across geographic locations. Why create additional risk?
We strongly recommend consolidating from steady-state to steady-state. After your consolidation is complete and you are operating in the steady-state for a period of time, we then recommend optimizing equipment and applications into a standardized architecture.
6. Business Case Not Formally Documented.
A data center consolidation needs to make sense for both your technology and business requirements. Your business case is a key tool to explain both and prioritize goals.
The objectives of the consolidation must also be clearly defined. Identifying the difference between soft and hard savings is particularly important. Developing and publishing metrics that measure progress and define success is instrumental.
We are all asked to do more with less. Emotions are subjective. But, black and white facts are hard to dispute. If you can’t measure it, you can’t manage it.
7. It’s a Part-Time job.
A data center consolidation typically takes 9-18 months to complete. A core team must be dedicated to the consolidation. Part-time resources cause gaps in availability (scheduling delays), ownership (wrong or incomplete plans) and unpredictable scheduling. Alternatively, full-time resources provide the continuity and corporate intelligence now and in the future. They are the force du jour, keeping the project on schedule and on budget.
In general, our advice is to build your team early with senior-level support. Involve HR from the get-go, too. A team lineup should include the more obvious technologists such as virtualization, network and storage staff, as well as many other more counterintuitive elements such as security, quality assurance, and procurement. Don’t forget a representative from each business unit who understands the interdependencies of their business applications and a representative from each vendor.
8. Stuck in Discovery
Discovery is a challenging process in every data center consolidation.
Most organizations are unsure about what data is important to collect. They face the dilemma of either missing critical information via incomplete discovery or wasting huge amounts of time and money via an exhaustive effort to collect absolutely everything.
How, then, does an organization accurately discover how many servers they really have? What are their dependencies and their business constraints? And, what process do you follow to ensure the right data is collected?
Most importantly, when are you “done”?
Experience allows a company to walk the perfect balance that avoids both severe risk as well as inflated cost. Discovery is strictly about fact-finding. Don’t allow yourself to get distracted and over-extended.
Your goal in discovery is to create a relevant, organized, living document called the Master Inventory List. Use it to capture every piece of data detail. It is essential for both accuracy and coordination.
A data center consolidation can, at times, feel like contained chaos. Something inevitably goes awry. And, the more complex the environment, the more potential there is for error. Your organization can mitigate the more common consolidation errors by following these recommendations. It is also important to remember that success is not the absence of consolidation errors but the ability to navigate them without disruption to services.
This certainly rang true for one of our clients who is a part of the world's largest marketing and corporate communications holding company. Their environment was complex, and it was exacerbated by dozens of their agencies independently operating with disparate technology, data security systems, and processes.