Data Center Outages and Risks

Nowadays, data centers are more interconnected and when you add cloud and increased data amount, you get a mountain of data moving in and out of data centers across the earth. Over the last decade the use of modern systems and processes improved the reliability of data centers. But customers are concerned about the growing risk of catastrophic failures, as more and more workloads and data are concentrated in larger data centers.

Uptime Institute’s annual survey shows that only 10% of the respondents experienced serious outages whereas it was 14% in the previous year (2019). While on-site power systems cause the majority of the power related problems, the root of the problem arises from the very systems meant to prevent them, such as UPS and ATS units. While power related problems remain to be the biggest cause of significant site outages, there are three other causes that stand out as particularly problematic;

• Cooling Failures
• Software / IT System Error
• Network Issues

These are some of the causes that remain at a high percentage. Reports show that failures of uninterruptible power systems, automatic transfer switches and generators are the most reported culprits of data center power outages.

About a fifth of all outages are classified as serious by the organizations that experience them. Research shows that cost of an outage has been steadily rising in the recent years. On average, over 40% of the outages cost more than $100,000 per hour. Since demand for capacity and digital services is expected to increase exponentially, the cost of any outage is expected to be higher in the coming years.

As the world becomes more dependent on IT services, reliability is going to require and will receive greater developments. Here at Boreas, we are developing technologies that proactively benefit your IT equipment and data center alike.