Continue in 2 seconds

Opinion 5 ways to establish dependable edge computing

  • January 11 2018, 6:30am EST

Edge computing is set to grow dramatically in 2018 and beyond. As IoT devices continue to come online in droves of tens of billions, edge data centers will grow in prevalence too, in order to collect, process and manage data when and where it’s being created. IT departments should expect to see tremendous growth and demand for reliable computing at the edge over the next few years.

As edge data centers continue to play an increasingly important role in both the business and IT landscapes, we’ll see the standards for their level of predictability and uptime grow to match those that enterprises and consumers have come to expect from traditional, large data centers.

So, how can you build an edge data center that’s reliable and generates value for your company? Here are five important ways IT departments should be building their edge data centers to help ensure end-to-end reliability and resiliency.

1. Ensure that your network is solid

We’re the most connected generation that ever lived, creating more data each day than any other time in history. As this unwavering connectivity permeates most aspects of business and everyday life, an edge data center is only as strong as its network. Solid core networks ensure that edge data centers are equipped to manage the rapid growth in demand from always-connected users and new data from multiplying IoT devices. Ensuring the highest levels of connectivity, availability and resiliency means redundant networking components, built-in link and gear failover capabilities, network security firewalls, and DNS and physical security.

2. Carefully consider your design and build

Despite their smaller size in comparison to enterprise data centers, edge data centers don’t necessarily guarantee significant cost savings. Considerations made in the design, build and operation stages of edge solutions should be the same as a large regional data center. This is because an edge data center outage can stall production and productivity, and even compromise revenue generation.

Equipment installed at the edge needs to be commercial quality with the same reliability features as those in larger data centers. Due to their distributed nature, repairs on edge data centers in the field can become far more complicated a lot faster than at a regular data center, and any outage may take longer to remediate due to a lack of onsite or nearby staff. Every edge computing component needs to be qualified and tested individually to make sure they work without fault before the setup goes online.

This is not to say every edge data center must be designed and built to provide the highest availability performance expected from an Uptime Institute Tier III or Tier IV data center. However, it does mean that the same disciplined approach companies take when deciding what availability will be needed, designing to that level of availability, and selecting and testing components for large data centers should be employed for edge data centers as well.

3. Prioritize remote management

As the term suggests, edge data centers are often geographically separated from regional data centers and the staff that supports and maintains them. So, it’s critical that edge sites be deployed in such a way that they can be tracked and managed remotely. Often isolated at distributed locations, edge data centers need to have critical capabilities like fault or failure recognition (both real-time and retrospective), remote notifications and alarms for off-site operations staff that are all accessible from a centralized command center.

4. Take extra steps to set up integral automations

For the same reasons remote management is indispensable, automation is equally as important to ensure a resilient and operational edge data center. Employ specialized building automation systems (BAS) to manage any edge data center environment, whether that be a manufacturing plant floor, a construction site, an oil rig or a small branch office. These systems will need to control the environment, report anomalies, and instruct failover to designed-in onsite redundancy when needed. A well thought out and high quality implementation with fully integrated automation is necessary for the data center to operate reliably.

5. Think carefully about IT infrastructure, resiliency, and physical space

Ensuring IT infrastructure can maintain both power and cooling at all times is an important component in the path to resiliency at the edge. Power infrastructure selection is vital to ensuring reliability and resiliency, whether it be redundant utility feeds or onsite power generation. Cooling is another key element. Know what your high and low load condition requirements are as well as the impact of all expected ambient conditions. You can then determine the appropriate cooling conditions for your edge operation ahead of your deployment.

Another important consideration that is sometimes overlooked in edge computing, is physical location. These data centers can be found under someone’s desk or even outside a warehouse, merely shielded from the physical elements. The sufficiency of an edge data center’s location (and the accompanying security implications) is completely dependent on your needs, but physical considerations should never be ignored.

At the end of the day, you should always think of edge computing setups as small but mighty data centers. Just because they’re smaller doesn’t mean the work they support isn’t valuable - or even critical. In order to ensure the reliability of your edge systems, take a pragmatic approach to their design, build and operations, just as you would a full-sized data center.

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access