5G Means It’s Time To Upgrade Your Database
By Andrew C. Oliver
What does a database company think about a major new infrastructure roll out? Well, that it is time to upgrade your database, of course. However, it has only a little to do with 5G.
A brief history of hype
First off, I can write the typical 5G article in my sleep: “What is it, 5G vs 4G, How fast will it be, lower latency, internet of things, something about self-driving cars that makes no logical sense, big avalanche of new data, upgrade your database.” In fact, so can everyone (and they have been). They can even recycle their 4G — or in some cases, 3G — rollout articles.
So far it isn’t the technology holding back a new generation of applications, it is U.S. mobile carriers and their data caps. Sure, they have “unlimited,” but I think we all learned a long time ago that the phone company doesn’t know what unlimited means. Luckily, they’re likely to relax their caps further when they update their network for 5G.
Secondly, it isn’t bandwidth that is holding back the Internet of Things, it is useful applications. I have a smart TV, thermostat, and some smart speakers. I didn’t buy a smart oven, dishwasher, refrigerator, or dryer. Moreover, my smart thermostat doesn’t generate a lot of data, and the only application — beyond finally being able to program it with my phone in a decent way — are the silly reports that Hitachi sends me that largely duplicate what Georgia Power sends.
Let’s get real though. Only one of these smart devices uses a lot of bandwidth: the TV. And it is really only a few applications, and they are all video. In fact, on my phone, most of my bandwidth goes to video and audio. Some of you might also spend a lot of bandwidth on games. That low latency is for you!
Finally, do you REALLY want a car that requires a constant internet connection to operate correctly? Do you REALLY trust any of the vendors who promise to deliver that connection to do so reliably? Naw, the self-driving car is largely speculative hype. Having more bandwidth will make it possible to update the AI models on the car more frequently and collect more data, but it isn’t really bandwidth holding self-driving cars back. It is a combination of consumer demand, poor road infrastructure and that the technology just isn’t there yet.
5G isn’t real yet
In fact, 5G isn’t real yet. The standards are still being developed and anyone selling you a 5G phone in the next couple of years is a huckster selling you a 4G+ phone that won’t be compatible with the 5G network when it is deployed (in the US that will be by anyone but Huawei apparently).
But what is real
More bandwidth is real. Assuming carriers lift their caps and the law of supply and demand are allowed to work. We’ve seen this before. The economic effect of 2 and 3G data was to dramatically reduce the use of normal voice calls. This led to a dramatic increase in the use of data beyond what the carriers could handle so they created data caps. They all tried to merge into one company, but when that wasn’t allowed competition from T-Mobile caused AT&T and Verizon to relaunch “unlimited” (which aren’t actually unlimited, they just raised the caps).
The biggest winner from 4G might have been Pandora and Spotify. The biggest winner from 5G might end up being Netflix (depending on how not neutral things are in the post-net neutrality era).
Higher data utilization is real. Video will certainly take the lion’s share (currently 83%) followed by audio. As for the rest, it may not be due to a slew of new IoT devices coming to your home or self-driving cars. Believe it or not, that’s a good thing! Uber, Amazon, Google and a few electronic manufacturers would have been the only companies that would have benefitted if those were real.
What is really going to happen?
Instead, data utilization will likely bump up a little with the rollout of 5G and continue on its current path. Every application will get incrementally better. Maybe Uber will show your car arrival a little more real-time, or maybe they’ll even REALLY show you how many cars are around. App developers who are holding back more streaming, real-time features will develop them knowing that users are not as worried about bandwidth as before and that their caps have been lifted.
So what does that have to do with your database? Why everything! There are going to be changes in your apps. There are going to be new sources of data, people are going to expect applications to be more real-time, combine more audio-visual experiences. Your old RDBMS already didn’t scale and wasn’t flexible enough to meet today’s demands. The NoSQL database you deployed because you heard it was cool, doesn’t even necessarily have the right architecture to scale to tomorrow’s demands.
Your traditional architecture doesn’t take into account the low-latency opportunities of edge computing and that sending data back and forth still uses a lot more battery power! 5G will just accelerate the curve of data utilization and user expectations, but if you aren’t using the latest, greatest database technology — you’re already behind that curve. It is time to catch up — if you wait for 5G, you probably won’t make it to 6G.
Andrew C. Oliver learned to code when he was 8. He founded the Apache POI project and served on the board of the Open Source Initiative. He writes a column for Infoworld and is the Director of Product Marketing and Evangelism for Couchbase.
