Is planning or implementing a data center consolidation/migration project giving you problems you need help with?
In an AFCOM Data Center Institute Study, 53 percent of U.S. companies responding expect to relocate or expand their data centers during the next several years. And nearly one-third say they will need to move, while 45 percent expect to make major improvements to their existing facilities.
I was asked by The DCIM Advisory to offer my suggestions to this. When I joined National City Bank, they were beginning the build-out of more data center space in two main facilities. The existing footprint consisted of many rooms built throughout the years. The latest technology at the time was used.
As with many data centers I had encountered in the past, the inventory management was based on spreadsheets and AutoCAD layouts of the racks and equipment. Three resources worked constantly to update and maintain the changes in the environments.
The asset center application that was the database of record had glaring gaps in what was reported in the facility, and the manual process as well had errors and inefficiencies. All rooms were filled to capacity, and it was clear that the build-out of new space was not the only change required to ensure that the business needs of the bank were met.
The consolidation and restack of the existing rooms became the priority post-construction to ensure that the proper balance of equipment to power and cooling was achieved.
Planning a data center consolidation requires knowing exactly what you have, where it is, and its configuration and connectivity. It also requires the ability to report and communicate all future plans to the appropriate individuals for approval and implementation.
Finally, a successful data center consolidation project is predicated upon having the solutions to effectively and efficiently manage and report on the status of the project throughout its duration.
There are eight steps that need to be followed in order to ensure a successful data center consolidation or migration:
- Poor planning is the No. 1 reason for a failed or over-budget data center consolidation and migration. Plan the capacity required to support the move, the time it will take, the skills required, and the interrelations between software applications and the systems that support them.
- Ensure that realty services and technology communicate. It is essential to have both groups on the same page and that expected outcomes are clearly defined. Data required for the proper execution of these projects exists in both of these realms and is essential for success.
- Establish a baseline to provide data for proper planning. Without the proper metrics, it is very difficult to track progress and success. You cannot manage what you cannot measure.
- Orchestrate changes during the move planning and execution periods to reduce risks to and complications of the project.
- Understand the needs and requirements of the business before planning and executing the project.
- Develop an accurate asset picture prior to starting a data center consolidation. Too often data center asset information is spread across the organization in spreadsheets, Visio diagrams and handwritten notes.
- Predict capacity, impact and expected results, setting goals for success.
- Include someone with data center move and consolidation experience on the project team. Many consulting organizations have individuals dedicated to projects of this nature. Ensure that the person you engage has a track record of success.
In the case of our needs, a DCIM solution was purchased and implemented. In approximately six months the system was successfully implemented. During the implementation of this system, we identified orphaned servers that amounted to 500KW of power and cooling capacity that was freed up for future growth.
Utilizing the features of the DCIM solution, we were able to identify unbalanced “hot spots” and adjust inventory to balance the load in the rooms.
Utilizing the modeling features of the DCIM system, we were able to identify the characteristics of the rooms and, in turn, the characteristics of the different types of technology equipment, developing a three-year life cycle restack to match a room’s capability with technology equipment that properly utilized the balance of space power and cooling characteristics. This added another 10-15 years of life to the facility.
Integration with provisioning, change, service and asset management systems ensures accuracy of the equipment and prevents future orphan events. The acquisition of the bank by PNC Financial Services provided another opportunity to model the equipment to the data centers of both companies, ensuring that the combined assets could be properly placed and consolidated.
There are new challenges that have developed over the past five years that have brought even more benefits to our having a DCIM system in place.
1. Budget cuts and the current economic conditions have reduced the amount of capital that can be expended on large scale initiatives.
a. C-level management now competes for a piece of a much smaller pie.
b. Success comes only from multiple C-level executives coming together on a given need or initiative.
2. Environmental awareness, energy costs, efficiency of operation and energy consumption are of global concern.
a. EPA issued a report to Congress addressing the effects that growing data centers are having on the environment.
b. Data centers have been identified as the top five industries for power consumption.
3. Facility environmental systems, controls and energy cost must be part of the total planning equation.
a. Retrofits of existing facilities and/or systems can yield capacity and longevity for the existing facility.
b. Physical plant capability matched to the technology that is to be deployed produces higher levels of efficiency.
Utilizing the open architecture and the integration potential of the DCIM system, we have been able to merge the data from our facility’s BAS and BMS systems providing the following:
a. A centralized dashboard and reporting of our data center environment.
b. Integration of real-time power data all the way to the plug strip.
c. Calculations of efficiency based on real-time data.
d. Energy management baselines to take advantage of energy rebate and utility company cooperative grid management.
e. The capability to calculate, manage and document the energy costs and environmental impact.
f. The reporting tools to support, trend and report on green initiatives or any potential future regulatory requirement.
This has placed the bank in a situation where it is prepared for the future in capacity and environmental management and possible regulatory reporting requirements.