This Industry Viewpoint was authored by Sharif Fotouh, Managing Director, Compass EdgePoint
If there is a song whose title characterizes the current state of the edge data center market, it would be The Temptations’ “Ball of Confusion.” (I’m a Motown guy.) Although you can’t swing a dead cat nowadays without reading or hearing someone describing “the edge” as being as close to data center nirvana as we’re going to get, right now it seems like we’re in a bit of “you go first” mode of adoption. Naturally, some may find this state of affairs a bit disconcerting, but if you think about it, this slow rate of adoption may, at least in part, be our fault.
While no one likes to look in the mirror and admit, “You know, maybe it’s me,” when it comes to edge computing, I think we need to start by embracing the adage “confession is good for the soul.” While we’ve been good about promoting the edge, we haven’t done the best job of explaining potential applications. Sure, we’ve convinced everyone that placing compute functionality closer to the customer is a good thing, but we haven’t done as well in answering the most basic of all consumer questions, “what’s in it for me?”
Perhaps we’ve been focusing a little too much on the “how,” when we should be emphasizing the “why.” In other words, applications speak louder than kilowatts.
Although IoT is still on the early end of the adoption curve, its fundamental requirements closely align with the benefits of edge computing. IoT applications depend on rapidly processing information to successfully performing a desired operation, thereby requiring low latency and the necessary amount of bandwidth to support the information flow. Reducing latency ensures the useful life of data is maintained, since many applications, like driverless cars, require “instantaneous” processing to perform as intended. A few milliseconds may not seem like a long time, but when it can mean the difference between safely completing your left-hand turn as opposed to hurtling into on-coming traffic, the concept becomes easier to grasp. As a result, potential edge solutions are more effectively positioned based on their alignment with the application requirements than their technology alone.
Although reduced latency is typically the element most often cited when discussing the benefits of edge implementations, including IoT, bandwidth may be the most important consideration. Digital traffic follows the same rules as its vehicular counterpart—the number of lanes (bandwidth) controls the volume rather than the length of the highway. In terms of bandwidth, edge data centers enable the application to be supported by multiple local high bandwidth connections as opposed to attempting to duplicate the same transmission requirements by “backhauling” traffic to a centralized facility. Note: A more in-depth discussion of “bandwidth economics” is warranted and will be the subject for a future column.
If we think about our IoT example, it isn’t unreasonable to say that the slower rate of edge adoption may relate more to our failure of imagination than we’d like to admit. In other words, we may be suffering from self-inflicted “when you have a hammer, everything looks like a nail” syndrome. For example, a possible use case for an edge data center would be to serve as a “mini-carrier hotel” to eliminate the connection costs and address the common problem of lack of available access to a regular carrier facility. In this example, the edge data center would be located near an existing carrier hotel to access multiple short-haul connections to a variety of providers. The edge facility then becomes an ersatz “meet-me-room” to provide a cost-effective mode of establishing carrier diversity. This type of application falls outside the standard use cases cited by, well, just about every edge data center provider.
The introduction of new technology or a technological concept is often greeted with unbounded enthusiasm while it’s only an article or a tradeshow presentation. It appears that the edge currently resides in that netherworld between a great idea and actual utilization. Eliminating the current ball of confusion will require providers to be more creative in identifying new use cases and a little less reliant on extolling the virtues of the hardware itself.
About the Author:
Sharif Fotouh is Managing Director of Compass EdgePoint and an ex-Googler. Fotouh is responsible for the Compass EdgePoint’s edge data center solutions as part of Compass Datacenters’ comprehensive core-to-the-edge offering to customers. He is recognized across both the information technology and the data center industries as one of the preeminent experts on edge computing. He has more than 10 years of tenure leading large data center and technology teams, including founding and leading Google Fiber’s national network facilities and deployment engineering program.
If you haven't already, please take our Reader Survey! Just 3 questions to help us better understand who is reading Telecom Ramblings so we can serve you better!Categories: Datacenter · Industry Viewpoint