Transforming Our Data Center with HPE Composable Infrastructure

You can’t do everything yourself. Even the most seasoned experts rely on trusted partners and solutions to transform their business. In fact, many of us in IT have experienced this as both an expert and a customer. It’s the mark of a healthy IT-industry ecosystem.

Just Fix It has been a managed service provider (MSP) since 2002, before MSP was a common industry term. From the beginning, our focus has been remote support as the first step in troubleshooting clients’ problems. We would put in place servers and small networks for businesses, and then problem-solve for customers as needed. 

Our business evolved significantly when Microsoft announced the discontinution their Small Business Server (SBS) solution, a product that we sold and supported often. Right away, we moved to a data center approach, which included hosting services and providing small business solutions by running the servers and software ourselves in a data center. 

Today, we serve a wide variety of customers in 10 different verticals, including global manufacturing, legal, and healthcare. All businesses have a core set of requirements, such as email and printing. But our customers also have industry-specific software for their day-to-day business.

Our challenge is getting to know those programs well enough to support them, while also providing those core required solutions in our cloud platform. We’ve become more like a specialized cloud service provider, hosting and supporting all of our customers' business applications in our private cloud in our data center.

Building Expertise: Growing as a Technician and as a Company 

I started off as an entry-level technician at Just Fix It when I joined the company in 2005. My background was in IT, building computers for a small computer shop. I came to the company with a good foundation but not a lot of experience in remote support. Experience came through training, but also through just figuring things out as I went along. 

Over the past 15 years, I continued to build my skills and experience and became more of an infrastructure manager, helping to design and build solutions for customers. Now as a senior IT manager, I oversee the infrastructure management and planning of our business as we continue to evolve.

As Just Fix It grew, I became increasingly concerned that we either wouldn’t be able to keep up with customer demands or that we would have a catastrophic data center failure. Our old infrastructure was made up of standard rack-mount servers and Storage Area Networking (SAN). Each of the servers and the SAN had their own monitoring and alerts component that worked well enough, but it wasn't all in one place. 

Because there was no central management, I wasn’t confident that it covered everything. If a port on a switch went down, we didn’t have that level of visibility to address it right away. We would know a connection was down—but we couldn’t know exactly where. 

That lack of visibility led to more site visits. We needed to have eyes on site, manually checking cables and other hardware to find a problem. We'd have to plan visits to our data center, which was usually about a half-day event because it is about 45 minutes away (depending on traffic!). On average, the visit lasted about four hours, and we'd probably do about 20 of those visits in a year, totaling around 80 hours on-site for various issues or equipment checks.

Discovering HPE Composable Infrastructure

Along with all the technical training and skills I learned on the job, vendor knowledge became increasingly important. I first learned about HPE through a project we were working on with an HPE partner, long before composable infrastructure came to the market.

For that project, a large manufacturer with its own IT department was implementing HPE Converged. We received training on it in order to provide frontline support, and I began thinking about what we could do with that level of infrastructure, consisting of blades, storage arrays, and fiber channel networking. We already had some cloud infrastructure at that point, but we decided to make it more robust.

Things stepped up a notch when I went to HPE Discover, HPE’s annual event. That year, I got to see the unveiling of HPE Synergy, the world’s first composable infrastructure platform. HPE Synergy technology would allow us to simplify our data center management and transform the way we frame, manage compute, storage, and networking in a single platform.

The timing couldn’t have been better. We had an aging cloud setup and were looking at the next generation of options so that we'd be able to scale up. I got in touch with our HPE rep about configuring a solution for us. 

A New Class of Data Center Infrastructure

HPE Synergy was essentially the first in its class on the market. From the start, I could see that composable infrastructure was a clear evolution of data center infrastructure. Combining disparate servers, storage, and networking into one package and the flexibility that it offered made the solution feel like a Swiss Army knife, something that embodied our remote support business model.

If we needed to change something around, I could shut down a blade, attach some storage, or modify a cluster without even getting out of my seat. At the time, that was not possible with any other solution.

Just Fix It was one of HPE’s first 3 Composable Infrastructure customers in Canada. We were confident that HPE Synergy would be a great fit for our needs, but since it was so new, it took some convincing for our management to sign-off on the solution. We needed to do our due diligence regarding other options on the market and the longevity of a brand new solution, so I did my homework and the team could see why I was excited about HPE Synergy above anything else.

Being a small to medium-sized business (SMB), we’re more nimble than most larger institutions. But the flip side of being an SMB is that we can’t afford to wait long. If something looks like a good solution and we get a good feel for it, we have to go for it as soon as we can. And that’s what we did by purchasing HPE Synergy.

Making Our Move: The Smoothest Transition for Us and for Our Clients

We moved equipment into a brand new rack of our data center. We set an install date, coordinated with the teams, and I met my colleague and four HPE technicians at the data center on the install date.

When we saw those huge crates of Big Iron, it felt like Christmas. It was such a new technology that working with the hardware was exciting for all of us. There were a few hiccups that required some extra troubleshooting, but we worked together to get it up and running quickly and then the HPE technicians handed everything off to our team. Internally, we'd already developed our plan of how we were going to deploy it. We then had to go through our own process of building up the hypervisor layer and storage virtualization to meet our needs. 

Our next challenge was to start transitioning customers from the old infrastructure to the new one. For the most part, we worked with our customers 9–5, and then we’d arrange for a weekend migration and implementation for anything we couldn’t do during their business hours. Email is a big client concern, and since we run an email on the service, mailbox migrations had to be handled with care to keep email flowing 24/7. But we did so many migrations, it quickly became old hat. 

We carefully transitioned customers, minimizing the impact to their business, and moved the old equipment out of production within about 8 months. We kept the previous set up running for training and development as a sandbox site. It's all out-of-warranty equipment that we'll maintain until we do our next equivalent refresh. Then we'll do a rolling refresh. 

Clients weren't quite as excited as we were about the new technology. They just wanted everything to work—and they were not disappointed.

Simplifying and Scaling

Once we deployed HPE OneView, we were able to automate our operations and simplify our infrastructure lifecycle management. It's a much cleaner rack, and beyond the physical equipment, everything is more streamlined. 

I now have much more confidence in the monitoring systems. I can log into HPE OneView and see everything there. Having that visibility is such a night and day difference. I can sleep better knowing that we have a much more robust infrastructure and have complete insights.

We've shifted our business model from being primarily a user support base to a service company. We’ve brought down user support needs and costs by providing better services through our data center and HPE equipment. This is where we see the industry moving as a whole.

Before, it was going to be a challenge to scale our old system. But our more streamlined infrastructure has changed our entire relationship with our data center. Where we once spent upwards of 80 hours per year at our data center over the course of a year, we’re now down to eight hours per year—a visit every six months or so. This has been particularly important during the COVID-19 pandemic, as we’re all doing our best to adhere to physical distancing protocols. Remote service and support has never been more crucial than in times like these.

Solid Support All-Around

One of the biggest differentiators throughout this process is when we’ve needed HPE support. Being in IT support myself, if I'd known the high level of support that HPE offers, it would've made the decision even clearer. I’ve had excellent experiences when I’ve called to work through any issues. When we’ve needed an HPE technician to replace a part, I provide them with security clearance, and they can access the data center without me needing to be on the premises.

HPE Synergy and HPE OneView have given us confidence that we can tackle any customer need. We've been able to bid on business opportunities such as data center migrations for larger businesses, that I’m confident we can handle smoothly. We wouldn’t have been able to say that before. 

Now, we know that we can handle growth—all because we knew when to trust the right partner to take us to the next level.