Keep up to date with the latest from Workspace Technology

Our Knowledge Hub Articles aim to offer readers additional insights on related technology topics. If you can't locate the information you need, please feel free to reach out to us, and we'll gladly address any questions you might have.

When AI and Data Centres Meet

23rd, September 2019

You’d be hard pushed not to work within a company that is data driven these days, in fact data has become an asset and a significant cog in the success of any business. Acquiring intelligence from equipment performance to customer purchasing trends, data is paramount to business growth and in some cases survival.

All this data needs to be securely stored and for this reason organisations are investing in state-of-the-art data centres to cope with this data demand. Having a data centre is one thing, but it also relies on professionally training personnel to maintain and monitor the day-to-day operations. Due to the expense of employing and supervising such specialist personnel it now calls for an alternative solution to the conventional method.

As some companies have already found, integrating Artificial Intelligence (AI) and Machine Learning into your data centre could be the perfect solution. A claim by Gartner states that 30% of data centres that don’t deploy AI and machine learning won’t be operationally and economically feasible by 2020.


AI brings a multitude of benefits which includes improving the security of your data centre. With the increasing threat of a cyber-attack, protecting your data is paramount. By deploying AI into your data centre, it can learn network behaviour and detect any deviation from the norm along with identifying malware and security loopholes in the system. It’s true that humans can do this too, but AI can process vast amounts of information much faster, and speed is the name of the game when dealing with the threat of cybercrime.

Energy Efficiency

Data centres consume a large amount of energy in order to operate which poses energy efficiency problems, with electricity in high demand, solutions need to be identified to solve the dilemma. AI is being used by leading organisations already for the purpose of conserving energy. Google have used AI to reduce their cooling systems' energy consumption by 40% which is the equivalent to millions of dollars for a company of their size. AI can learn and monitor temperature set points and flow rates to collect critical data with the help of smart sensors which assist in designing or configuring a fit for purpose data centre. This approach can identify sources of energy efficiencies and automatically fix the problem reducing overall consumption.

Preventing Downtime

Manually predicting data outages can be a complicated task relying on humans to analyse different elements to find the root cause of a problem. AI on the other hand can monitor server performance and predict data outages and implement mitigation strategies to assist the data centre to recover from an outage.


Monitoring software is already available; however, AI takes the capability a step further for active equipment monitoring tasks using Deep Learning (DL) applications to predict failures ahead of time. AI can identify defects in data centre equipment through pattern-based learning. By using sensors installed within the equipment, excessive or low vibrations and odd sounds could be detected and sent to the engineers to investigate further helping to predict potential equipment failure.

Reducing the Head Count or Improve Productivity

AI and machine learning can also cut down or even eliminate the number of staff required to operate your data centre. If cost cutting is not the aim, then it will certainly allow employees to concentrate on more strategic and rewarding tasks rather than focusing on the continual monitoring of tasks.

Microsoft’s underwater data centre initiative ‘Project Natick’ is an example of removing the need for human intervention. It has been designed to function for five years without any maintenance, by using AI it screens for abnormalities that could result in equipment failure.

Capacity Planning

AI can assist in forecasting demand so that the data centre does not run out of power, cooling and even space. A data centre is a significant investment and understanding how many servers you require, the amount of cooling needed could save not only on the pocket but also your time. With the increasing amount of data, traditional data centres will eventually get slow and result in an inefficient output.

AI and machine learning are expected to optimise every aspect of data centre’ operations, including the planning, design, maintenance and control. By 2022, IDC predicts that 50% of IT assets in data centres will be able to run autonomously because of embedded AI functionality. This is clearly the future of data centre management, but are you ready? As data storage increases and infrastructure gets more complex, relying on human decisions and intuition alone is not going to deliver the level of accuracy and efficiency required.

< Return to Blog