With the last year’s “lockdowns” due to the COVID-19 pandemic, keeping data center systems operational to support increasing numbers of remote workers, supply chain partners and critical customers has been the top priority for every business around the world. From healthcare providers to financial services firms to content and media providers, our customers on Platform Equinix® have relied on us to help them serve their users around the clock across the globe, without disruption.
Over the last year, we have leveraged some tried and true best practices, especially in the area of fire prevention and suppression to increase the operational integrity and safety of our data centers.
System and UPS redundancy is still the most reliable way to avoid IT downtime
With remote access to corporate IT services becoming mission critical for remote workers over the last year, we’re still adamant about UPS usage to ensure resiliency for our customers’ business systems.
In an Uptime Institute survey of 300 data center managers, they were asked “How the pandemic would change their operations?” Two-thirds expected to increase the resiliency of their core data center(s) in the years ahead.”[i] Because most businesses are operating with fewer staff onsite, the possibility of a system failure, even with automated remote monitoring and preventive maintenance, has increased. To lower the impact of a system failure, you need to ensure power redundancy.
At Equinix, we provide fully redundant UPS power systems and dual power circuit feeds connected to two UPSs to our global data center customers. In order to take full advantage of this redundancy, we recommend connecting your equipment and/or your Power Distribution Units (PDUs) to both circuits.
We also see many companies putting more than 50% load on one of the circuits, making it impossible for a single circuit to take on the full load in the event of a failure. For the best results, balance the load on each circuit so that if one fails, the other one can manage 100% of the power requirements.
Network redundancy is now the top priority
According to the Uptime Institute survey, network issues were cited as the most common reason for any IT service outage — more common even than power problems.[ii] In fact, 39% of survey respondents said, “they had experienced an outage caused by a third-party networking issue.”
Network redundancy is pretty common with Equinix customers given the more than 1,800 network service providers (NSPs) in our expanding ecosystem. However, connecting to multiple NSPs can contribute to additional complexity and lack of visibility. We’ve helped our customers solve for these issues via Equinix Fabric™ software-defined interconnection across our global platform. There are two physical ports on Equinix Fabric that you can rely on for additional resiliency when connecting to NSPs, cloud service providers (CSPs) or your own IT infrastructure on Platform Equinix. Because Equinix Fabric interconnects digital infrastructure on our global portfolio of Equinix International Business Exchange™ (IBX®) data centers across various metros and geographies, you can create interconnected business continuity and disaster recovery scenarios that best meet your business needs. To learn more about Equinix Fabric and geo-redundancy, read our article about geo-resilient scenarios for protecting mission-critical workloads and data and connecting to the cloud.
How to protect against fire risks
Though very rare, data center fires do happen. According to Uptime Institute’s database, which documents over 8,000 abnormal incidents since 1994, only 11 fires in data centers have been reported by the institute’s members — less than 0.5 per year. Most causes of IT fires are from human error —overloading power strips or accidentally starting fires by putting a flame near highly flammable materials. Today there are different types of automated power shut-off and fire suppression systems that will detect smoke and heat and stop flames from spreading. For example, lithium-ion batteries have been known to be a fire risk in many smart phone and laptop systems. To mitigate fire risk, the battery performs internal monitoring at the cell level, which cuts the battery power if heating occurs.
Uptime Institute provides the following guidance to reduce your data center fire risk:[iii]
- Use VESDA (very early smoke detection apparatus) systems and maintain appropriate fire barriers and separation of systems.
- Keep well-maintained water-based or low pressure clean agent fire suppression systems.
- Perform risk assessments, primarily aimed at reducing the likelihood of system/power outages, to identify issues with these systems.
Prevention is the best precaution
As the world’s digital infrastructure company™, we are proud of our industry-leading track record of >99.9999% uptime in our 220+ Equinix International Business Exchange™ (IBX®) data centers. We incorporate innovation in our global IBX data center infrastructure designs and interconnection services to ensure the highest degree of interconnectivity, security, reliability and efficiency. In these challenging times, we maintain our deep commitment to service excellence and work closely with our customers to deploy best practices that keep their mission-critical systems up-and-running and safe, at all times.
[i] Uptime Institute, “Why data center operators are investing in more redundancy,” October 26, 2020.
[ii] Uptime Institute, “Network problems causing ever more outages, ” February 22, 2021.
[iii] Uptime Institute, “Data Center Fire Frequency,” April 5, 2021.
With remote access to corporate IT services becoming mission critical for remote workers over the last year, we’re still adamant about UPS usage to ensure resiliency for our customers’ business systems."
Only 11 fires in data centers have been reported by Uptime Institute’s members since 1994 — less than 0.5 per year.