Easing Management Burdens and Increasing Efficiency with HCI

Before the coronavirus was on anyone’s radar here in the eastern United States, 2020 had already dealt us a bad hand.

In January, the conversations among the IT team were all about a potential malware attack. What would it look like? Were we prepared? Our data center had recently been through a couple of incidents involving power outages, so we were pretty well versed in disaster recovery (DR) protocol. But in February, we suffered a major malware attack, with multiple servers and a lot of data encrypted with ransomware. It was our worst nightmare, come to life.

This was pretty out of the ordinary for a company like PITT OHIO, a transportation and logistics provider. Less than truck load (LTL) shipping has been our traditional core business since the company was incorporated 40 years ago, but recently we have transitioned more into logistics and supply chain. In short, we will touch any type of freight that goes on a truck to service our customers, either directly or in-directly.

I climbed up the infrastructure team ladder here over the last 15 years and am now the director of EIOS (Emerging Infrastructure and Operations Support). Looking back on it, I honestly didn’t think I’d be here this long. But despite our growth, PITT OHIO has retained the company culture that first drew me in. It maintains a fairly flat management structure and has the family-owned mentality that values people above processes.

Moving Toward Hyperconvergence

That being said, growth always brings challenges. In our case, our staff growth wasn’t keeping pace with the overall growth of the business, which meant that some of our admins wore a lot of hats. Instead of having a dedicated storage admin or a server admin, for example, we had people who were each responsible for dozens of tasks. 

Additionally, we had a lot of separate best-in-class solutions. Each solution worked well for its specific function, but it was hard to maintain everything—and maintain a high level of expertise about everything—with such a small team. We’d previously implemented Cisco UCS in 2009, and it made a big difference from a management point of view, but it was still more complicated than we would have liked. We knew we wanted to go further and explore hyperconverged infrastructure (HCI).

The opportunity arose in 2014, when I came across a small booth for HPE SimpliVity (just SimpliVity at the time) at a Gartner conference . I made a note of the company and kept tabs on them for the next year. The other competitor at the time was Nutanix, but we liked HPE SimpliVity’s stated premise: to solve the I/O and data predicament.

That spoke to us, because our biggest issue has always been storage. We were constantly running into capacity and performance issues, always trying to squeeze all these resources on a shared load on the SAN. Whatever solution we chose had to solve those problems, and that’s what HPE SimpliVity promised from the beginning. 

With HPE SimpliVity, we could potentially consolidate our knowledge and tech stack into one and make it more efficient, both in terms of management and also financially, due to less cabling and hardware. We moved forward with a proof of concept, and, satisfied, decided to go all in on the solution.

We spent roughly 1.1 million for the initial build out. We saved a little more than $200,000 compared to our traditional approach, and that savings included both capital and expenses.

Growing Together with HPE SimpliVity

HPE SimpliVity was new on the scene, but we took a chance on them and our companies have grown together. Even early in our talks with HPE SimpliVity, we engaged directly with members of the executive team, who always took us seriously and wanted to hear from us. Our relationship has been strong since the beginning, and that connection has only gotten stronger over the years.

When we first started dealing with the company, they weren’t owned by HPE. They were small, meaning that they lived and died by their customer service. We didn’t run into many issues, but it was nice knowing that we’d always talk to the same people whenever we opened a request. It was also nice knowing the leads from the technical, product, and marketing teams, who constantly followed up with us and asked how we were doing. That feedback loop meant that my team helped develop features that were later included in product roadmaps. 

Our pre-HPE SimpliVity DR compared to our current DR scenario is like night and day. We previously used a Data Domain for offsite backups, so we had our data sitting somewhere else, but it didn’t have extra compute. With HPE SimpliVity, you basically build out DR as you build out your environment. It’s all the same experience, in the same environment, and the redundancy and replication is built in. As long as you buy enough hardware and put it on a secondary site, you have backups for everything. 

Before, backing up was a nightly process. Now, our data is everywhere. If I need to recover anything from those nodes in our secondary site, I can do that very easily. 

We saw that firsthand during our malware attack earlier this year. We had been running one legacy system that we hadn’t switched over to HPE SimpliVity. Of course, that was the system that got infected, so we couldn’t recover everything. Everything that was backed up on HPE SimpliVity, however, was very easy to recover. No doubt about it: If we didn’t have HPE SimpliVity, this would have been a complete disaster. It saved our bacon.  

Another feature we’ve watched develop is HPE InfoSight. We used this solution before HPE bought Nimble, so we were already familiar with the technology before it was rolled into HPE SimpliVity. InfoSight made Nimble different from other providers in the SAN space, so we were excited when HPE explained how it was going to play into their overall cloud strategy. Right now, we use it for asset tracking and predictive intelligence; it tells us that a drive may soon fail, or that we’re nearing storage capacity. 

This is a good way to manage our environment, but the most exciting aspect of InfoSight is the potential for intelligence. One of the biggest things we’ve always encouraged with HPE is the push for more automation. InfoSight is a great start. It’s already made us much more efficient because we can set it and forget it. We've seen efficiency improve even more by incorporating additional predictive intelligence into our growth projections and having the ability to move resources around.

Performance and Scalability

HPE SimpliVity turned out to be a game-changer in terms of human resource management. We no longer need staff with specific expertise to manage and maintain our technology; the simplicity of HPE SimpliVity makes it much easier to find and retain talent that can handle everything.

Regarding performance, one of our clusters has a deduplication ratio of 50.2:1 and the compression is 1.4:1, which I think is pretty standard for those who use HPE SimpliVity. The real story on HPE SimpliVity’s performance is how easy it was to add capacity to our VDI environment when COVID-19 hit us in March. 

I remember meeting with our CIO on a Monday, when I was told to start thinking about ways to send everybody home and retain productivity. That Tuesday, the directive was, “We need to send everybody home now.” That meant we immediately had to set up 250 people for remote work. 

Luckily, we already had a big VDI environment on HPE SimpliVity, so enabling additional remote workers was just a matter of adding some Citrix servers without even having to add any hardware to the environment. That’s primarily because of the shared storage and the way the I/O is handled. The way HPE SimpliVity does the data dedupe makes a big difference, and allowed us to stretch to what we needed. Within three days, we had more than 150 people working from home. In the past, there’s no way we could have scaled that quickly.

Gaining the Inside Track

HPE SimpliVity still reaches out to us to get our feedback, including to beta test new features. In fact, I think that has gotten even better since the HPE acquisition. With the launch of their Insiders program, they share plans with dedicated users, and I can see how the information we share with them actually gets incorporated into their features roadmap. 

HPE Insiders is also an avenue for users to talk to one another and ask for advice. We’re getting ready to do VMware NSX and one of my team members, Justin Brooks, who’s developed a lot of great relationships through the Insiders program, was able to find somebody who has experience running NSX on HPE SimpliVity. We can learn from their knowledge and experience, and others can learn from us. That peer support is invaluable. 

We’ve done the same for other companies. This is probably the best environment I’ve experienced to share with other users directly, rather than going through a vendor sales team. It’s not too intrusive in the normal course of your workday, but it’s always there when you need it.

I can’t say whether HPE SimpliVity is right for everyone. Every organization is different and has different needs, so due diligence is a must. What I will say, however, is that if hyperconvergence is right for your business, then HPE SimpliVity should be in the running. Nothing’s ever perfect, but it has huge potential to make a difference for your team. 

It’s the gold star among all the experiences I’ve had with technology providers, and the results speak for themselves. It’s all that and a bag of chips.