According to a study conducted a few years ago, our patience threshold is declining in the Digital Age. Three years ago, the average amount of time spent a user would wait for a web page to load before “losing it” was 10 seconds. And, how long were we willing to wait for a video to buffer before the ole’ blood pressure started to go cray-cray? 16 seconds. And that was back in 2015!
It’s no wonder then that a “holy grail” of IT operations is achieving low cost, low latency application performance. Especially, when it comes to one of the fastest growing technology segments today: The Internet of Things or IoT.
This practice of embedding sensor technology in all manner of electronics is predicted to exceed mobile phones as the largest category of connected devices in 2018 according to a recent study conducted by Ericsson Mobility. Correspondingly, in addition to the insane growth projected for IoT data, more and more of this IoT data processing is getting pushed out to the edge to get as close as possible to the source sensors and end-users of the resulting data analytics.
So, what’s the key to success in designing low latency IT infrastructure solutions? Well, to coin a phrase commonly used in real estate, “Location! Location! Location!” The shorter the physical distance between sensors, data analytics applications and the end-users of the processed data, the better the experience.
Now, when it comes to making infrastructure design decisions to optimize latency and cost, there are a lot of variables to consider. But, here’s the bottom line: the more location options you have, the better.
To put this into context, let’s think in terms of planning a travel itinerary from your corporate headquarters on the east coast to a remote sales office on the west coast. To keep travel time to a minimum, you’ll want to:
Choose the Shortest Distance
Fly into the airport that is closest to your destination. If you have a choice between a major airport (think public cloud provider like AWS here) that’s 100 miles from the office and a smaller, regional airport that’s 10 miles from the office (think private Colo from an edge datacenter provider here). The math often points to the small regional airport as the optimal final destination to save on drive time, mileage and fuel costs. Besides, who wants to drive 100 miles to get to work?
Have an Abundance of Convenient Options
Let’s take this analogy one step further…
If you’re fortunate enough to have multiple regional airport options, choose one that offers:
Scheduled, non-stop flights to and from your major airport hub to save on flight time and minimize layovers.
This is the equivalent of selecting localized data center providers with multiple cloud on-ramps that are direct-connected to public CSPs.
Multiple rental car company options that are on-site to take care of your last mile needs
This is the equivalent of choosing a location that offers software-defined networking (SDN) infrastructure with access to multiple connectivity providers.
An online presence in the most commonly used travel reservation sites such as Expedia or Travelocity so that you can easily coordinate ticketing and reservations with multiple airlines, airports, rental car companies, hotels, etc. through a single user interface.
This is an emerging market for which there are no clear leaders yet. However, QTS’ Service Delivery Platform (SDP) offers the ability to manage hybrid cloud environments and coordinate provisioning across multiple networks.
As you plan these expanded edge routes for other remote offices, an interconnected mesh of both major hubs and regional airports will begin to emerge (think hybrid cloud and colo networking here). This mesh will give you the shortest path from point A to point B and require the least amount of drive time and last mile fuel expense (think connectivity charges here).
These are concepts that are dominating IT conversations these days, where we refer to them as multi-cloud, edge data centers, SDN networking and colocation networking.
Now, back to our travel analogy, what if you wanted to establish a trial period to compare flights, airports, traffic patterns, etc. across multiple options and providers before making any long-term decisions such as leasing a private company jet and hangar at the airports with the best locations? In these situations, it’s best to choose a regional airport that offers charter services that do not require a long term financial commitment. That way, you can test out different routes before pulling the trigger on that shiny new company jet.
So, in summary, if low latency and rapid application response time is important to you, whether you are deploying IoT workloads or not, then consider taking advantage of next generation, hybrid cloud service providers that offer on-demand colo, multi-cloud on-ramp services and software defined network connectivity options that are a short distance from your final destination. The right combination of these cloud and colo service providers will let you push resources out to the edge and bring the datacenter to your customer instead of bringing your customer to the datacenter. Because time is money and as Twitter co-founder Evan Williams said, “convenience decides everything.”
By the way, if this blog post does not load in under 10 seconds, you are excused from reading it.
Contact us to learn more about QTS Hybrid Colocation solutions.
QTS Product Development Manager