Leveraging AI for efficient and resilient data centres

Schneider Electric - According to an S&P study, the Asia-Pacific region will grow at a faster rate than the rest of the world between 2017 and 2022. It will reach an estimated 10 per cent CAGR compared to the global data centre industry that is expected to clock in at 7 per cent CAGR.

With the huge population of young and tech-savvy consumers in the region, this surge is unlikely to let up soon.

Building better data centres

As more data centres are built, it should come as no surprise that organisations are looking at how they can enhance their infrastructure with cutting-edge technologies to reduce cost.

Around the world, the pandemic has forced enterprises that relied on in-house support personnel to realise that they might be more vulnerable than they realised to situations beyond their control - such as government-mandated lockdowns or travel restrictions. 

Fortunately, technological advancements mean that these challenges don't have to be show-stoppers. For a start, a new generation of cloud-based management platforms makes it possible for IT to support both core data centres and edge deployments remotely and safely.

Vital data centre operations can be monitored with greater efficiency than before, with potential failures automatically flagged to keep unplanned downtime and outages at bay. 

Moreover, cloud-based Data centre Infrastructure Management (DCIM) can help enterprises ensure the reliability and availability of their infrastructure from afar.

With the requisite visibility and secure remote access to connected infrastructure, reliability can be heightened with new micro or edge data centre deployments.

AI for predictive capabilities

Elsewhere, the use of artificial intelligence (AI) within data centres has also made great strides forward.

With ample streams of real-time data from digitalised systems, algorithms that leverage historical data can deliver actionable recommendations to optimise infrastructure performance and maintain uptime.

Finally, predictive algorithms can inform data centre stakeholders of impending disruptions before they happen.

Indeed, it is now possible to build even more reliable data centres with AI-powered predictive capabilities.

And by repairing a data centre asset just before it fails, the US Department of Energy concluded in a study that predictive maintenance is more cost-effective than preventive maintenance in the form of regularly scheduled maintenance by between 8 per cent and 12 per cent.

For comparison, reactive maintenance - or not performing any maintenance until it breaks, is 40 per cent more costly than reactive maintenance.

Taken together, an organisation that switches from reactive to predictive maintenance can significantly decrease their cost while increasing reliability.

By integrating the cloud and AI, enterprises can boost their efficiency in terms of energy use and operations, as well as enabling greater resilience. There is no question that AI will underpin the future of data centre infrastructure management.

As data centres continue to evolve, expect cutting edge technologies to dovetail with people and processes to build even more resilient, cost-effective data centres. 

Read more about how Schneider Electric is leveraging AI to drive greater sustainability in this interview here.