Andreopoulos on Acropolis Part 1

A disclaimer before we get started. First, if you are expecting an epic “battle of the hypervisors” article, then please move on. This is going to be a multi-part series as I attempt to explore the viability of AHV (Acropolis Hypervisor) in the market place for a variety of use cases. Secondly, this is a long article, and in this first part, it’s mostly a history lesson in IT technology paradigm shifts.

Just because I have recently made to move to Nutanix, doesn’t mean I have short term (or long term) memory loss about the value of the solutions I have evangelized for the last decade. In part 1, I’m going to take a little trip down memory lane and provide some history and views from my perspective; having spent over 20 years in this industry in a variety of roles…both from working in IT, to a vendor capacity. In fact, it’s for this very reason why I decided to join Nutanix, which will hopefully become clear after reading part 1.

My career in IT started about 23 years ago, when the world of IT centered around a little company known as Novell. Remember these guys?! Yeah, well I may be older than some of you reading this post, but let me start by saying that idea of becoming a Netware CNE was a dream sought by many back in these early days. It was the launchpad that allowed me to move on to bigger and better things in my career. The main reason I’m starting with Novell, is to paint the theme that I’ll be addressing here today , and help everyone remember what is easily forgotten.

I was working as a system administrator for a large home builder that was primarily comprised of Netware servers, running on HP infrastructure, in the mid to late 90s. If you recall, there was a battle brewing in the industry between the NOS giant Novell, and the desktop software giant Microsoft. My fellow CNE rat pack would chuckle when we saw our peers attempting to run their file/print, and application servers on NT/Microsoft. There was a rather eccentric IT director of Applications (whom I’m still friends with to this day), that was almost single handedly responsible for bringing Microsoft and NT into our organization. We were cocky….We thought our #(*%&@* didn’t stink. Needless to say, 5 years later, we were the ones being laughed at by the MCSE/NT fan boys (sorry girls).

History lesson #1, time to re-invent myself and my IT career. This was the first, and not the last time, something major and game changing would have such a huge impact on me, my career, and the industry as a whole.

One thing I took extreme pride in, was how we built our datacenter (s) infrastructure. We had a cookie cutter reference architecture that we replicated at all of our home offices. It was simple, elegant, and repeatable. If we ran into issues, we were already 75% way to root cause because there were no layers of complexity to navigate. Pro tip of the day: remember what I just said when we talk about the subject of complexity in a later article. So we recently had our heads bashed in by Microsoft, and in comes our wonderful CIO (seriously though, he was one of the best CIOs I’ve worked with) asking the question “Umm folks, I want us to take a look at Dell servers.” We gasped, we chuckled, we pushed back. LOL, Dell servers really??! He is joking right. Let’s jump ahead about 3 more years and guess what? If you walked into our shiny new datacenter in Tempe, AZ, you would see nothing but racks and racks of that Dell midnight gray color everywhere. Not a single HP server in sight. We know that history repeats itself, but in this case, wow, it happened again, and it happened quickly.

Ok, granted, this wasn’t as game changing as the Novell catastrophe, but it was quite an impact. We were reluctant, and we didn’t believe it was possible that Dell could play in this space. We were wrong…again.

Lesson #3 was by far the most impactful, but this time I was on what I consider to be the beneficial side of the fence. We were a classic IT shop with the “one app, one server” paradigm. Our internal customers (application teams), were hogs and selfish. They didn’t want to share, and they wanted it all. After all, why not? They didn’t have to pay for anything 🙂 A couple admins on our team were investigating a little company called VMware, and were pitching the idea of “virtual” servers…

Ok, spoiler alert, but everyone knows what happened next right?! To me, VMware has probably had the most profound impact on our industry for over a decade now. We, like every other large company, transformed our datacenter through virtualization. We saw our footprint shrink, we saw our sever vendors scramble, we saw the advent of the SAN and shared storage, and we saw entire careers and skill sets take shape through this new industry shift. This time, the history lesson was for our internal customers. They had to play nice in the sandbox with everyone else, but in the end, we all reaped the benefits.

Before I touch on AHV, I wanted to give an honorary mention to Cisco and the release of their UCS server platform. I wasn’t particularly impacted by this, but I will say, many of us were skeptical (and chuckling inside) about Cisco entering the x86 server space. Once again…we were WRONG! History lesson #4.

I think by now, you should have a good understanding of where I’m coming from here. After living through so many game changing technology paradigms, I’ve come to pretty much expect these will happen today. A big part of my decision to make a career change recently was because I started having that same feeling I experienced over my career so many times; that there must be a better was to do this. I’ve had such a great time in recent years, helping customers solve problems through automation and orchestration, but in the end, I couldn’t help but have that feeling that there must be something fundamentally wrong with the way we have approached IT infrastructure.

I’m not going to provide a pitch on Google, Facebook, and how the “webscale” approach to IT is so game changing. There is plenty of data available for your consumption out there. I will say this, however, that is the fundamental reason for my career change. I’m convinced that this new architecture is so game changing, that this just might be History lesson #5. If I had to build my own datacenter today, I would certainly be considering all options and looking at ways of simplifying our approach. Heck, maybe I would just throw in the towel and put everything in AWS!?

The Nutanix platform comes with a built in hypervisor built on KVM known as AHV (Acropolis Hypvervisor). Is this thing for real?? Should I start “chuckling”. How good can it be? Ok, these are questions I will attempt to answer in future parts of this series, but for now, based on my limited experience with the platform, it appears to be quite legit. One of the things I absolutely love with AHV is the ability to manage it in the same UI (Prism) as the rest of the Nutanix platform. There is of course some very core functionality that any hypervisor must have, and I’ll finish with quick synopsis of the feature set. We will get in depth into bits and bytes in later parts of this series:


Ok, now the final point of my article. I’m a huge VMware fan boy. The platform just simply rocks, it works, and everyone has it. Are there viable alternative hypervisors out there today? If you would have asked me that question 6 months ago, I would have “chuckled” yet again. Does that mean I was wrong? I’m not going to attempt to predict the future here, I’m only going to ask that you think about the history lessons I’ve outlined above. You decide what the future holds, but we should all be prepared for changes in every part of our industry. They will continue to happen and I’m just happy to be along for the ride. The more you learn, the more you…

One thought on “Andreopoulos on Acropolis Part 1”

Leave a Reply

Your email address will not be published. Required fields are marked *