"Practical Guide to E nsuring Availability at Edge Computing Sites"

Practical Guide to Ensuring Availability at Edge Computing Sites

White Paper: Schneider Electric

Deployments for distributed IT are typically relegated to small confined rooms, closets, or even on the office floor. However, as businesses grow and rely more on edge computing applications, IT downtime has a greater impact on the business. An interview with a small food distributor exemplifies this sensitivity. As this food distributor gained more customers, they realized it wasn't possible to fulfill orders accurately and on time without their IT systems. Downtime of these systems would not only interrupt their distribution schedules, but would cause restaurants to place last minute orders. Restaurants need only a few missed deliveries to have a good reason to seek a new distributor. The following are some example causes of downtime that were uncovered in this research:

• The wrong server was unplugged. The IT admin thought he had traced the correct power cord to the tower server. The "rat's nest" of power and network cabling significantly increased the likelihood of this error. Dual-power supplies later became a standard specification for critical IT gear to avoid this type of human error.

• A server error for high temperature forced a shutdown of the system.

• A few pieces of IT gear turned off during a brief power outage. It was later discovered that the equipment was never plugged into the installed UPS. This was most likely due to the disorganized cabling behind the rack.

• A cleaning person unplugged a server to plug in the vacuum cleaner.

• A power outage caused all the systems in a branch office IT rack to go down. The IT admin arrived later to discover that the UPS had been signaling for some time that it had a bad battery that required replacement.

As with many businesses, especially small businesses, it takes a downtime event or a series of close calls to finally invest in improving the availability of IT operations. In many cases, this spurs new IT upgrade projects. An upgrade project is the optimum opportunity to assess the physical infrastructure required to support IT, however, our research suggests that IT managers often lack the time to research and specify an appropriate solution and the plan for deploying it. We addressed these two needs in two papers. This white paper provides practical guidance on how to improve the IT availability at these sites, while White Paper 174, Practical Options for Deploying Small Server Rooms and Micro Data Centers, describes a practical plan for deploying a micro data center at one or more locations. Furthermore, you should perform a health assessment to ensure old equipment is modernized as discussed in White Paper 272, A Framework for How to Modernize Data Center Facility Infrastructure.