← All Posts

The End Of The Cloud Is Not Coming

Written by
Kyle Galbraith
Published on
7 November 2017
Build Docker images faster using build cache banner

It’s Saturday morning and I am scrolling through my various feeds when I see this article from Venture Beat, “The end of the cloud is coming”. Now baited, I take a sip of my coffee and click through.

Disclaimer: If you do believe the end of the cloud is coming, move along because you’re not going to like this post.

With that out of the way I will go ahead and argue the reasoning used in this article. The author states four different reasons why the cloud is going to end. Let’s go ahead and talk about each one and why the reasoning is completely ridiculous.

It can’t meet long-term scaling requirements.

The argument the author uses here is a bit mind blowing to be honest. Networking is the reason the author believe cloud providers like AWS, Google Cloud, and Azure can’t meet the long-term scaling needs. We are pushing unprecedented amounts of data through networks.

As servers move to the cloud, i.e. onto Amazon’s or Google’s computers in Amazon’s or Google’s data centers, the networks close to these places need to have incredible throughput to handle all of this data.

It is true we are sending a lot more bits around networks nowadays. But the idea that networks close to Amazon or Google data centers can’t handle the load is an uneducated answer. A bit of research or watching a re:Invent video or two and you see that AWS leverages their own network.

The author goes on to say that CPU’s and hard drives will be at capacity to serve content to every user. Uh has someone never heard of caching? Streaming optimization? Clearly not.

It’s centralized and vulnerable.

This is your usual regurgitated argument about how you are storing all your data in a centralized place. A nice scare tactic in here as well:

What if Amazon’s data center gets flooded, hit by an asteroid, or destroyed by a tornado?

As of September of this year every major cloud provider, AWS, Google Cloud, and Azure have multiple regions and availability zones. So if you are concerned about a zombie apocalypse attacking your data center like the author believes, replicate to multiple AZ’s and even regions for the North Korea hydrogen bomb as well.

It demands trust but offers no guarantees.

Passing data through a middleman like AWS you are implicitly trusting a middleman. If you are passing a half baked blog post from your laptop to your contact at VB, you are implicitly trusting a middleman.

The difference? Passing data around an AWS network you are trusting Amazon to handle your data. Surfing the web, you are trusting your ISP as the middleman. Given the current state of Net Neutrality, which one do you prefer?

It makes us — and our data — sitting ducks.

Yup services are collecting loads of information on us. Unfortunately we are all to quick to hand over details about our lives to play Angry Birds or Pokemon GO. It is unfortunate that technology insists on making money on information when there is better ways of doing so.

That said, data sitting in cloud provider data centers is secure. But if the author wants to have data sitting in their own data centers like an Equixfax data center in Georgia, who am I to say no?

Putting The Myths To Rest

Alright counterpoint arguments and rant over. I do believe that the author is trying to get to a bigger point. The point is that a peer to peer network could become the future over the traditional data center. It is nice in theory and there is a lot of movement on this front with blockchain technologies.

Gartner magic quadrant

The cloud is not going anywhere anytime soon though. The magic quadrant makes that pretty clear above with AWS far out front. The real question to ask is how will the cloud evolve next? Microsoft and Google are fighting hard to rundown Amazon. It begs the question what are the next set of innovations that push things forward?

© 2024 Kyle Galbraith