Originally posted on The Independent Data Center Alliance

There is no question that next-gen technologies like artificial intelligence (AI) and automation are having a massive impact on enterprises globally. Consider, for instance, that by 2022, companies are expected to have an average of 35 AI projects in place, and 86% of CEOs say AI is mainstream technology used in their companies today.

Data centers also are being significantly impacted by these technologies, from both external and internal applications. As global enterprises increasingly utilize such technology, demand for data centers increases exponentially, as well. In fact, the data center market, which currently is valued at about $32 billion, is expected to grow to $59 billion by 2025.

According to Bluebird Network’s Todd Murren, General Manager of Bluebird Network Underground, he is seeing more managed service providers installing bare metal servers to host applications, particularly for artificial intelligence.  “These customers are running GPU’s, graphics processing units, which are typically found in gaming machines,” he says.  These types of processing units “require massive amounts of power and produce tremendous amounts of heat…that can have a substantial impact on data centers,” Murren added.

And, these technologies also are being used by data center operators to impact data center management, productivity and infrastructure, including energy optimization, identifying defects and predicting equipment failures and even staffing.

In fact, almost 80% of data center managers say AI will be used in their facilities to manage staffing levels – and 34% of those say it will happen by 2025. That’s not surprising considering that AI has the potential to impact data center operations in four key ways:

  • Power management: Such solutions optimize heating and cooling systems, which can cut electricity costs, reduce headcount and improve efficiency
  • Equipment management: Technology is used to monitor the health of servers, storage and networking gear to ensure they’re properly configured and predict when equipment is about to fail
  • Workload management: Solutions can move workloads to the most efficient infrastructure in real time
  • Security: By learning what normal network traffic looks like, solutions can spot anomalies, prioritize alerts, help with post-incident analysis and provide recommendations to improve security

To read the full article please click here.