J. Clayton Tallieu

Follow

Feb 25, 2017 · 5 min read

I was introduced to my definition of cloud in early 2007. It was around the time Amazon released Amazon Elastic Cloud Compute or EC2 in 2006. A good friend and colleague dropped by and said, “Hey Joe, let me show you something.” Whenever Vijay said this, you stopped everything and gave full attention, because he was about to crush your belief system or send you off doubting your own existence. He was in the limited public beta, and introduced me to EC2.

At that time, this was a real game changer for a developer with an entrepreneurial spirit. Before EC2, there was a slight barrier for someone to create a new service. You had to build hardware, host it somewhere, pay for enough bandwidth and maintain those things. It was a real investment to stand up a service on the internet. Most everything I did up to this point in my side-hustles were isolated experiments contained in my lab.

Then EC2 came… Now we could have a fully networked server with beefy compute power on a whim. Wanted to try a thing? Go get a new EC2 instance and build it. Don’t like it. Destroy it. You paid for as much as you used. This meant the initial cost barrier to try something was really really low. You were not heavily invested.

The capabilities provided by the cloud expanded and other players entered the cloud market. You had Rackspace, Microsoft Azure, Digital Ocean and other big players like Dell offering compute in the cloud to run your software. You had PaaS atop those providers, like AppFog and DotCloud (makers of Docker) giving away ready made LAMP stacks. And the value adds kept coming. MongoDB in the cloud, Stripe for secure credit card payments, and on and on.. It was trivial and virtually free to cobble together a web service around 2011. I know, because I did just that.

Building A Service On the Cloud

In 2011 over the 2-week Thanksgiving for a total of $44, I built an inventory managment system bridging Point of Sale and E-commerce systems on DotCloud for a monthly fee. But, I would not consider this a cloud service. It was a service built ON a cloud. I did not build a cloud.

I know from experience, like myself, many companies have been able to capitalize on a small investment by building IN the cloud...stitching together other SaaS platforms. Some eventually turning them into a successful business or a useful service. As a consumer of these services, you are once removed from the underlying cloud supporting this service and are a bit exposed to the risks that your provider has accepted on your behalf.

Building A Cloud

When Eagle Eye Networks reached out to me about working on their middleware API, I was intrigued not only because cloud video surveillance sounded sexy, but because it seemed like they were HEAVILY invested in their product. Doing a little research on Dean Drako made it clear that this was no ordinary Austin-type start-up. I didn’t fully comprehend the difference, but my “spidey sense” was tingling, and I wanted to find out. I was in.

In my early days at Eagle Eye Networks, the way they did things was at odds with what I was used to. And I was naturally resistant and skeptical to some of the practices. I could not comprehend why a company in this age would not use AWS or Rackspace. Or use Amazon S3 for storage? It seemed unnecessarily costly to build it all yourself. Then it became very clear to me. They are BUILDING A CLOUD, not building IN A cloud. They were making that serious investment I mentioned earlier on behalf of it’s customers who are serious about security. We are in the security business.

You had to build hardware, have it hosted somewhere, pay for enough bandwidth and maintain those things

The important distinction between building a cloud and building on the cloud became evident twice since I started at Eagle Eye Networks. The first was the world-wide DDOS attack that brought down the internet in Oct 2016.

Amazon Web Services was also experiencing connectivity issues on Friday around the same time as the Dyn attacks. AWS is used by more than 1 million companies, including GE (GE), News Corp (NWS). and Capital One (COF).

We didn’t notice a thing. Except that, because we ourselves use cloud services, our development and business operations were halted. But the service we provide to our customers did not blink.

— — — —

The second instance was just yesterday, when Cloudbleed became a new term.

Have you heard? A tiny bug in Cloudflare’s code has led an unknown quantity of data — including passwords, personal information, messages, cookies, and more — to leak all over the internet. If you haven’t heard of the so-called Cloudbleed vulnerability, keep reading. This is a scary big deal.

Guess what? We are the cloud. We do not use Cloudflare, so our customers are not impacted.

Summary

When choosing to use a cloud service it’s important to understand how these services are built to properly assess the risk. Making the decision to rely on the cloud can be scary.

Is the service you are choosing, built in the cloud, or are they building the cloud? Eagle Eye Networks is building its own cloud and does not leverage other cloud services in its product. This is an important distinction to understand when selecting a cloud video surveillance company. From the decision to build the cloud to the architecture of the edge devices, it’s all about cyber security at Eagle Eye Networks.