Data center maintenance and operation audits: when, why, and how to conduct them?

As cornerstones of organizational information systems, data centers represent an asset that must be maintained and improved over the years to achieve the expected levels of service continuity, security, and energy efficiency: this is the entire objective of actions undertaken in terms of operation and maintenance.

Whether performed internally or externally, these actions involve regular milestones. Many contexts can thus lead to a data center maintenance and operation audit, such as a change in on-site personnel, the more or less regular appearance of performance or even availability issues, the inadequacy between installations and evolving needs, the desire to reduce operating costs, or the completion of an assessment – on what worked well or less well – during the renewal of the operation and maintenance contract.

In all cases, a maintenance and operation audit aims to measure the differential between what is expected and what is actually achieved, to define potential levers for improvement, and to provide concrete elements for defining or redefining priorities.

The maintenance and operation audit is a very comprehensive procedure covering contractual, procedural, and of course, technical aspects. It consists of elements:

• of analysis (room audit, inventory of technical equipment, operation and maintenance procedures and protocols);
• of existing evaluation (maintenance audit and contracts with various service providers, evaluation and testing of their interventions, identification of weaknesses for security, business continuity, and recovery in case of incident);
• of proposals for potential areas of improvement (recommendations and improvement programs, estimation of achievable gains: service quality, reduction of operating costs, etc.).

It is carried out in two phases, one of which can be described as theoretical and the other as practical.

Phase 1: Documentary study

To gain complete visibility of installations, procedures, and interventions, a data center maintenance and operation audit begins, on the one hand, with a study of the As-Built Documentation: sizing, site operation, correspondence with functional requirements (Tier III…). And on the other hand, with the analysis of operational documents: procedures implemented on site (access, escalation, quick reference guides…), interventions (work orders, compliance with Guaranteed Intervention Times (GTI) and Guaranteed Restoration Times (GTR)), monitoring of energy consumption and spare parts stock, or significant events (personnel changes, on-call duties, breakdowns, replacements, main corrective and preventive tasks performed, quotes issued by service providers, etc.).

The purpose of this documentary study is to ensure that operation and maintenance have been carried out according to industry best practices and in accordance with what was planned, over a given period. Thus, missing or insufficiently documented records over time can signal a certain lack of rigor in the execution of dedicated maintenance and operation tasks, which implies a deeper investigation during the on-site visit.

Phase 2: Technical site visit

Regardless of the results of the documentary study, an on-site technical visit is essential. The purpose of this visit is to ensure consistency between documentation and existing conditions (operational equipment and completed interventions), to carry out a complete inventory of equipment (electrical power systems, climate control, UPS, etc.), and their condition (unloaded elements, dirty filters, etc.).

Generally, this technical visit should also be an opportunity to understand the additions of hardware elements made progressively (why, how, for what needs met, what maintenance?), to assess the supervision tools and alarms in place, and to concretely evaluate the corrective and preventive maintenance interventions performed (delta with the content of the work orders).

In conclusion, the information gathered during the audit should enable organizations to better understand the costs associated with their data center and to make the right choices regarding operation and maintenance, in order to improve performance and extend the lifespan of their equipment under the best possible conditions.