According to Gartner, more than 75% of all enterprise-generated data will be created and processed outside of the traditional data center or cloud by 2025. Instead, much of this data will be handled by edge data centers.

Edge data centers are smaller facilities located close to the populations they serve that deliver cloud computing resources and cached content to end users. They typically connect to a larger central data center or multiple data centers. By processing data and services as close to the end user as possible, edge computing allows organizations to reduce latency and improve the customer experience.

Examples of edge data centers include modular data centers fabricated from steel shipping containers, prefabricated edge pods, small stand-alone brick-and-mortar data centers, phone-booth-sized enclosures that hold a single rack, and very small enclosures that only hold one or two servers.

The growth in edge computing can be attributed to transformative technologies like online shopping, Internet of Things (IoT), artificial intelligence (AI), content streaming, 5G, blockchain, remote learning, and telemedicine. As the volume of data from such services grows, there will be an increased demand for more edge data centers to process and store that data locally. However, these edge sites are often near outdoor or semi-controlled environments that threaten equipment health and uptime. Data center managers must follow edge infrastructure management best practices to maintain edge site health and uptime.

Best Practices to Ensure Edge Site Health and Reliability

In a recent American Society of Heating, Refrigerating and Air-Conditioning Engineers (ASHRAE) technical bulletin, the association shared advice to edge infrastructure managers to highlight environmental and reliability challenges of managing edge data centers. Here, we’ve compiled some of the best practices included in that bulletin:

• Understand IT equipment warranties before the deployment. For each piece of IT equipment you plan to deploy at an edge site, you should carefully check the manufacturer’s warranty to understand the allowable temperature range, rate of change of temperature, humidity range, dew-point limits, dust filtration, and air quality requirements. It is important to know the conditions that apply to the warranty because opening the edge site door may expose equipment to outside conditions that void the warranty and compromise the equipment.
• Design the enclosure to support the narrowest of IT equipment warranty ranges. Avoid exposing IT equipment to outside air that may violate the conditions of manufacturers’ warranties by designing an enclosure that will maintain the allowable temperature, humidity, and air quality ranges.
• Mitigate risk during equipment service. If accessing equipment within an edge site could expose it to outdoor conditions and exceed temperature or humidity warranty limits, consider rescheduling the service for a different time of day or until conditions improve, using a mantrap or airlock to minimize environmental shock to IT equipment, or using a temporary environmental containment device with a heater or A/C unit to enclose the door of the edge site enclosure and keep conditions within the specified ranges.
• Monitor temperature and humidity during service. Technicians should monitor both the temperature and humidity of the IT equipment inlet air in real-time while equipment service is being performed. This will help you know if you are exceeding warranty limits or ASHRAE guidelines. This can be done with a portable handheld unit, a readout from equipment firmware, or with environment monitoring software.
• Don’t forget about air quality and corrosion. It’s easy to focus on temperature and humidity, but air quality and corrosion must be considered for edge sites that may be exposed to outdoor conditions. Gaseous pollution can cause equipment failure from corrosion of exposed copper and silver on components and printed circuit boards. Dust and lint can clog inlet air filters and foul heat sinks, diminishing airflow, increasing fan power draw, and reducing cooling efficiency. Be mindful of the location of edge data centers as proximity to coal-fired plants, agriculture, strong winds, ocean spray, and even heavy vehicle exhaust during rush hour can all release dust and pollutants that damage equipment.
• Keep an eye on energy efficiency. The increasing trend of decentralized data centers has implications for energy efficiency and sustainability. These small sites are often less efficient than their central counterparts and can lower data center fleet energy efficiency and increase operating expenses. The small footprint and reduced insulation can disproportionately increase the thermal impact of outdoor conditions while small enclosures may reduce airflow.
• Be aware of reliability and redundancy challenges. The reliability of an edge site may be impacted by its location due to workload proximity, environmental factors, power reliability, and longer mean repair times for remote locations. Similarly, systems may not be fully redundant. Fewer IT equipment pieces in an edge site means there is a larger impact if one piece fails, more redundancy is required to account for longer service times at remote locations, and it may be more difficult to physically separate redundant equipment. An uninterruptible power supply (UPS) and backup cooling system may be needed for unstaffed and remote sites.
• Deploy Data Center Infrastructure Management (DCIM) software. According to ASHRAE, “the disparate nature of edge data centers make the role of remote management and monitoring not only central, but ever more critical, to their operational performance. Tools such as DCIM play a key role in ensuring that all the relevant parameters can be remotely monitored.” DCIM enables edge infrastructure managers to remotely monitor temperature, humidity, IT equipment health, physical intrusion, water intrusion, fire, and other data from environmental sensors and intelligent equipment.

Improve Edge Infrastructure Management with DCIM Software

DCIM software enables you to centrally manage all resources and capacities across all global sites in a single pane of glass to maintain uptime, improve the efficiency of capacity utilization, and increase the productivity of people.

Deploy a second-generation DCIM solution to:

• Get the most out of space and power resources. Since edge data centers are much smaller than traditional sites, it is important to maximize the utilization of available capacity. DCIM software provides color-coded floor map reports so you can easily see and correlate common capacity constraints like cabinet weight, percentage full, budgeted power, and actual power. What-if analysis can be conducted to understand the potential impact of installations and decommissions on space and power capacity, enabling you to identify stranded capacity and defer capital expenditures or know that you need to purchase additional resources.
• Issue visual work orders to remote hands. Moves, adds, and changes must be completed accurately and quickly, but directing technicians on what to do can be difficult when you’ve never been to the edge site. With DCIM, 3D floor map visualizations of each edge data center allow you to see where all your cabinets are located, and rack elevation views let you see the exact U-position and other details of your assets. You can show technicians exactly where data and power ports are located so that connections are made correctly.
• Monitor data center health of all locations. Get at-a-glance views of the health of every data center and edge site with an enterprise dashboard that shows real-time power and environmental health and events for all sites in a single pane of glass with the ability to drill down for cabinet-level details. This information lets you be the first to know of potential issues that can cause downtime such as power capacity limitations, hot spot formation, or loss of redundancy.
• Accurately track and manage all assets, parts, connections. Organizations with edge data centers typically have complex and distributed deployments spanning many sites and multiple business applications. DCIM makes it easy to keep an accurate, real-time inventory of all assets like servers, networking equipment, rack PDUs, patch panels, cabling, and parts and spares. You can track metrics like asset count by application, hosts per application, and asset cost by location to know which applications demand the most resources and see where you can improve efficiency. Visual circuit trace diagrams enable you to avoid overloading circuits, decrease latency, and quickly troubleshoot connections to reduce downtime.
• Protect all sites and assets. Remotely restricting physical access to edge data centers can be difficult. DCIM provides the ability to remotely control electronic cabinet door locks, restrict access with role-based permissions, ensure compliance with real-time audit logs, and monitor video surveillance feeds to maintain the physical security of your edge sites and equipment.

Bringing It All Together

Maintaining edge site health and reliability can be a challenge when you have no onsite personnel, no visibility into what’s happening, and potentially harmful environmental conditions near IT equipment. However, by following best practices and using the right software tools, you will improve uptime, efficiency, and people productivity at all of your sites.

Want to see for yourself how DCIM software can help you manage your edge infrastructure? Test drive Sunbird’s second-generation DCIM today.