Follow us on

FREN

Data Center Relocation: No Room for Improvisation

Published on June 3, 2020 -

A data center relocation requires meticulous preparation, both organizational and technical, to minimize or even avoid service interruptions. From defining the relocation scenario to restarting servers, what are the steps to follow?

Data Center Relocation: Defining Target Strategies in Advance

Relocation, replacement of an aging IT room, new hosting scenario, IT infrastructure rationalization, or outsourcing: numerous situations can lead an organization to decide to relocate a data center. Just like an office move, relocating an IT room cannot be improvised, must be prepared well in advance, and involves both IT teams and business departments (accounting, production, logistics, etc.).

Data centers support an increasingly significant portion (if not all) of business operations. Therefore, before even considering any scenario, the very first step is to identify the contacts capable of providing all the necessary information to carry out this project while minimizing, or even avoiding, service interruptions.

The first step is to establish a precise mapping of applications, their interdependencies, their criticality level, potential downtime windows, etc. Then, to conduct an exhaustive inventory of equipment (servers, racks, etc.) associated with running these applications. This assessment also makes it possible to evaluate actual requirements in terms of electrical power and cooling capacity for the target infrastructure. Based on criticality constraints and possible downtime windows, the team in charge of the relocation can then establish priorities and decide on a complete relocation or a phased relocation by batches.

Relocating an IT room, since all equipment will be removed from the racks, is also an opportunity to completely rethink its layout: reorganization of racks, power and data cabling, PDUs, and cooling equipment. This step also makes it possible to determine which equipment to retain or replace.

Ultimately, only by taking all these parameters into account will it be possible to design a coherent scenario, according to the selected target architecture, and based on downtime constraints and priorities.

Operational Preparation: At the Heart of the Relocation Project

After this initial phase of defining the relocation scenario, which takes between one and three months depending on the complexity and size of the infrastructure, the fieldwork begins, which will lead to project success: operational preparation.

This begins with establishing a provisional schedule, taking into account various constraints and the possible existence of a business continuity plan, defining whether or not it is possible to proceed with relocating different batches, as applicable, on business days, working days, weekends, or at night. Not to mention the necessary cooling and rest times for equipment before and after the relocation (which can each require up to one hour) and, of course, the actual transport times, depending on the distance to be covered.

In very concrete terms, while relocation to a hosting provider generally involves simply ensuring consumption limits per rack based on weighted server consumption, moving to a dedicated facility amounts to nothing less than designing a new data center from scratch. This requires special monitoring in terms of organization and governance: operator, consulting firm, IT services company, etc.

Everything must be studied and planned: the moving company, the number of trucks required, the sites (access procedures, existence of a loading dock, freight elevator, etc.). And, on the future IT room side, cross-connections, cables (type, length, color, naming), equipment labeling, network equipment, PDUs, racks, cooling units, hot and cold airflow in compliance with best practices for layout and operations, etc. All while taking into account the technical characteristics of the various IT equipment, according to manufacturer standards.

Data Center Relocation: A Flawless Plan?

Provided the previous steps have been scrupulously followed, the physical relocation is undoubtedly the simplest part of the project. Each identified piece of equipment must be able to almost instantly find its place (and connections) within the new data center. Objective: this step must be limited to executing the plan defined in the preparatory phase, to minimize the risks and uncertainties that could arise.

Once the "Go" is given, machines are shut down, and cooling times are observed, equipment must receive special attention and be perfectly protected from the risk of drops or shocks. The use of flight cases and specialized transport companies is particularly recommended in this regard.

Despite all these precautions, it is advisable to take out ad valorem transport insurance: no one is immune to the unexpected, and IT equipment replacement costs are significant enough not to take unnecessary risks. This insurance is offered as standard by companies specializing in data center relocation.

In all cases, the success or failure of a data center relocation project depends on the project management skills of the team, internal and/or external, responsible for it. It is preferable to rely on employees or a third-party company with solid experience and/or certifications capable of demonstrating this expertise.

Our services

APL Data Center offers a range of services covering the entire data center lifecycle, from strategy to operations. Each service is part of a comprehensive approach designed to ensure performance, resilience, compliance, and sustainability for critical infrastructures, whatever the technical and operational challenges.

Consulting - Data Center Strategy

In an environment where data is at the heart of business performance, every hosting decision has a long-term impact on your organization. APL supports companies and hosting providers in defining robust, scalable data center strategies aligned with their business challenges.

Design & Construction - Engineering

Designing a data center means orchestrating technical, energy, regulatory, and operational constraints, while guaranteeing a high level of availability.
Engineers working in data center rooms.

Do you have a data center build, transformation, or optimization project?

Would you like to learn more about the benefits of digital twins?