Scale logo white
Contact
Edge Assessment
Pricing
Demo
edge computing deployments

4 Tips for Edge Computing Deployments

by Dusty Koekenberg • Oct 27, 2020

Edge computing deployments have unique constraints that are significantly different from the ones with which typical data center deployments contend. After all, by definition, edge deployments are away from normal support services, far from the sanitized data center, and deeply enmeshed in the real work of the organization where they must deliver high value without disrupting other business activities.

So, organizations pursuing edge deployments should step back and carefully consider all the different factors that must be weighed for edge computing and then look for a solution that addresses the most crucial. Let’s take a look at four of the most crucial.

Tip 1: Keep the Footprint Small

It might seem an obvious point, but many products sold as “edge” equipment are not actually designed for that purpose—they’re simply marketed for that role or, at best, minimally adapted for edge requirements.

Because some vendors just sell standard data center equipment for edge use without accounting for the less-than-perfect environment that may be encountered there, edge installations are acquiring a reputation for being unexpectedly troublesome.

That’s why it’s vital that edge equipment should be conceived from the start for that purpose, with sufficient ruggedness built in to handle the wider range of issues that are typical at the edge, whether that edge is the shop floor, a warehouse, or somewhere in a retail establishment.

Edge components and systems need to be thought of as “universal” products that can be deployed when and where needed, with few limitations, and made appropriately secure in any given environment.

Tip 2: Autonomous and Resilient

Edge computing is no place for daintiness: It’s where real work gets done, some of it dirty, messy, hot, and noisy. So, in addition to needing only a basic physical environment and simple power and connectivity, prudent adopters make sure the edge setup (hardware and software) is designed to be failure resistant, able to recover from many problems autonomously, to protect data, and to maintain operations in almost any circumstance.

What does that look like in practice? As much as possible, physical interventions onsite need to be avoided—no buttons that say “press here” to restart something. Instead, everything should be callable and able to respond when power is made available. Furthermore, autonomy should be part of the basics, delivering no-nonsense reboots and allowing most other maintenance tasks to be initiated remotely.

Tip 3: Easily Scalable and Expandable

Edge environments are dynamic, with new applications being deployed regularly and data volumes growing exponentially, creating new demands on edge infrastructure. It’s critical that this infrastructure is designed to accommodate that growth and expand and upgrade the edge micro-data center with new resources and applications as easily as the initial edge deployment.

Failure to plan for expansion of the edge environment can lead to expensive forklift upgrades or multiple independent islands of infrastructure to manage, with all the complexity and cost associated with that kind of choice.

Tip 4: Standardized and Simple To Manage

For all but the smallest of organizations, this is perhaps the most important consideration because edge may involve multiplying sites and types of equipment on the network.

If approached haphazardly, without a plan, edge can quickly spawn hard-to-manage complexity that can strain IT staff and have company-wide implications. To keep from becoming a nightmare, edge systems should take a standardized approach requiring little or no customization and minimal skills in installation.

To that end, edge should embrace infrastructure as code (IaC), which simplifies change control.

Repeatability means that service and support is standardized so staff doesn’t need to research each installation before responding to a problem; instead, use a consistent approach and methodology. It’s a model for efficiency used across every other domain, from manufacturing to medicine, but too often ignored in edge deployments.

Similarly, management must not require specialized IT staff on site; upgrades and infrastructure scaling must be non-disruptive; the foundation must be self-healing; and IT specialists must be able to manage the entire edge fleet seamlessly at scale.

Finally, look for zero-touch provisioning. This is a device-configuration process that can be operated automatically and eliminates most of the burden on IT administrators when setting up, maintaining, or upgrading an edge system.

If you’re interested in learning more about our Edge Computing solutions can transform your IT infrastructure, start your free trial today!


Scale Computing Is A Gold Sponsor at the 2020 Acronis Global Cyber Summit

by Dusty Koekenberg • Oct 08, 2020

Read More

HC3 Backup Appliance - Limited Time Offer!

by Dusty Koekenberg • Nov 11, 2020

Read More
Scale Computing

Contact Us


General Inquiries: 877-722-5359
International support numbers available.

info@scalecomputing.com
Solutions Products Industries Support Partners Reviews
About Careers Events Awards Press Room Executive Team

Hoosier Pride and Silicon Valley Innovation

Privacy Policy – © Scale Computing 2023