AI brings perfect storm to data center operations
AI to address 3 common human/operations related problems
Gartner predicted that by 2020 more than 30 % of data centers that failed to sufficiently prepare for AI would no longer be operationally or economically viable. Well, we’re 2021 and it has become clear that AI, when deployed strategically and paired with adept human oversight, can generate a host of new efficiencies for next-generation data centers and can address at least these 3 human/operations related problems in today’s data centers. In addition, the energy and carbon savings produced by AI/software-driven data center optimization solutions will help data center operators on their net zero journey towards carbon neutrality in 2030 (as specified by the EU Green Deal).
1. Prevent data center outages by predictive analytics & early warning capabilities.
According to Uptime Institute research, outages are becoming bigger and more painful but more importantly, 75% of data center operators admit most outages were caused by human error and were preventable with better management/processes or configuration. With only 5% of data center managers gathering the real-time and highly granular data they need to effectively manage the thermal performance of their data center, it’s no surprise that at any given time 10-15% of data center racks are well out of ASHRAE thermal compliance. AI and smart sensors can greatly assist here for monitoring & compliance purposes.
2. Mitigate skills & staff shortage.
The shortage of skilled candidates in the data center industry is a problem that has been around for many years and is now nearing a crisis. According again to Uptime research, 50% of operators has difficulty finding qualified candidates for open jobs causing them to recruit outside the industry, retrain staff or turn to outsourced resources. Choosing an AI assisted hybrid management model, where AI tools support operators with some of the management, monitoring functions, provides a happy medium between outright automation and chronic understaffing.
3. Remote monitoring of critical sites during Covid-19 pandemic.
Maximizing data center performance is difficult enough during normal conditions, optimizing while also trying to accommodate the sustained added pressure due to the Covid crisis is challenging. Having visibility of thermal performance at critical sites becomes ever more important, certainly at a time when many data centers can’t get their teams physically on-site due to lockdown challenges. Remote thermal monitoring helps data centers manage their facilities during the Covid-19 lockdown.
Feel free to contact us if you want to discuss how AI tools can assist your data center operations and improve the sustainability in your data center at the same time.