Posted on

Why audit your data centre?

The purpose of an audit is for an independent third party to examine the data centre and assess it against best practice to identify where it is compliant and where improvements can be made.

However, the purpose of the audit is not always the same. Here we will identify some of the more common reasons for carrying out a data centre audit.

Tier, Rating or Class Compliance

The most common reason for auditing a data centre is to assess the Tier, Rating or Class of the data centre.

This raises the question, “Which standard do you want the design to be certified against?”

Some of the key data centre standards referring to Tier, Rating or Class are;

  • ANSI/TIA-942 Telecommunications Infrastructure Standard for data centres (Ratings)
  • EN 50600 Information technology. Data centre facilities and infrastructures (Classes)
  • The Uptime Institute Data Center Site Infrastructure. Tier Standard: Topology (Tiers)

Not all standards have the same interpretation of Tier, Rating or Class. The main aim of classifying a data centre design in terms of Tier, Rating or Class is to establish its resilience to failures. The aim of a higher Tier, Rating or Class is to remove single points of failure from the design and to enable maintenance to take place on the infrastructure without the need to shut down all or part of the IT systems (often referred to as concurrent maintainability).

It is worth asking…

Do I really need my data centre certified?

If you are building or operating a colocation data centre then you may want to certify your data centre against the TIA942 or EN50600 data centre standard. Provided you use an independent, competent and experienced consultant (like Capitoline for instance 😊) then your customers will have the confidence in their statement of compliance. The certificate of compliance can then be used as a marketing tool.

On the other hand if you are building or managing your own data centre then it is important to understand the vulnerabilities in the design but the need for a formal certificate of compliance is questionable.

What most data centre owners, operators and users are really interested in understanding is…

Do I have any Single Points of Failure in the design?

Will I have to shut down my IT equipment in order to carry out routine maintenance on the data centre infrastructure?

You can find a detailed comparison of the main data centre standards and about data centre certification in our white paper

Troubleshooting and Failure Analysis

Reliability is perhaps the most important requirement for a modern data centre because of the dependency organisations have on IT systems. Assessing the Tier, Rating or Class of the data centre, as we have already discussed, can help us to mitigate against failures.

But if the IT systems fail it is a major issue and so we must respond appropriately. The first priority is to get the IT systems back online.

Why did it happen?

How can we stop it happening again?

The first question may be easy to answer. Perhaps the cause of the IT systems outage was the failure of a component in the power or cooling systems. If so we can fix it and get our systems up and running again but if we do not invest time in answering the second question then we may suffer the same fate again. It may not be the same component that fails but perhaps we have an inherent weakness in our data centre infrastructure. This should be investigated, the weak points identified and remedial work be recommended. The management team can then make an assessment as to whether the cost of the remedial work compares favourably to the reduced risk of another failure.

It is important to understand that not all failures are caused by the physical infrastructure, many are caused by operational issues. We must therefore also review operational processes and staff experience and training to ensure that we have minimised the risk of a repeat failure due to these factors.

Energy Efficiency Audit

A typical data centre consumes 50 times more energy per square meter than an office building. Data centres are responsible for more than 2% of the worldwide production of Carbon Dioxide. Carbon Dioxide has been linked to global warming and so it follows that we should be trying to reduce energy consumption in data centres.

Concern for our organisation’s “green” credentials and being conscious of environmental issue are important but often the main driving force for reducing energy consumption in data centres is that…

Improving energy efficiency in the data centre saves money!

Not only does improved energy efficiency reduce running costs but it can also increase the data centre’s capacity to house IT equipment. A more efficient data centre can make more power available to the IT equipment.

There are many ways in which the efficiency of a data centre can be improved, some are low cost and very simple to implement, others require a long term view to be taken.

Recommendations for reducing energy consumption in data centres can be found in EN 50600-99-1 Information technology. Data centre facilities and infrastructures. Recommended practices for energy management

Operational Policies, Processes and Procedures

It is becoming increasingly recognised that solely focussing on data centre design to achieve high reliability in data centres is a mistake…

Data centre operations management is just as important as design!

To prevent data centre failures many organisations spend millions on redundant equipment and alternative power supplies. However, if you do not have the right documentation, operational processes and suitably trained staff that money can be wasted.

A Tier 4 data centre with poor operations management can behave like a Tier 1 site in terms of reliability!

Good operations management is not just about putting maintenance contracts in place. We must also have…

  • Up to date documentation
  • Appropriate organisation structure
  • Trained staff
  • Appropriate policies
  • Appropriate operating procedures
  • Planned and corrective maintenance
  • Monitoring systems
  • Measurement and feedback of KPIs (Key Performance Indicators)
  • Business Continuity and Disaster Recovery Planning

An appropriate standard to compare to is the EN 50600-3-1 Information technology. Data centre facilities and infrastructures. Management and operational information.

You can find out more about this standard in this article…

Or alternatively learn about this in detail in our DCOM Data Centre Operations Management course.

We have looked at the three most common reasons for carrying out a data centre audit but others include;

From experience of auditing many hundreds of data centres we have found no two customers have exactly the same requirement and most audits are a combination of the requirements discussed here.

If you would like to know more about data centre auditing then please take a look at our DCAF Data Centre Auditor – Facilities Course.

Related Courses