What is our primary use case?
We have 30 compute nodes and about 34 storage nodes that we're trying to consolidate down in a co-location area. We use it for all of our workloads outside of our main EMR system. It does everything from Oracle, SQL, all sorts of workloads, CIFS, and NFS.
How has it helped my organization?
HCI definitely improved how flexible we scale, and our entrance into the cloud. The features are very rich, in terms of both avenues. It's helped us flexibly move and shift our workloads around, back and forth.
Application performance has improved.
It has helped reduce maintenance costs. Our old footprint was just huge in terms of the environment we need to run it, it was very bulky and old. It increased all costs. Even just operating the equipment, but the maintenance of it and how many people we need to take care of things has been reduced. The daily maintenance seems a lot more consolidated down to something easier. We had a team more of around 10 to 20 people doing everything. It's consolidated down to about seven or eight of us but it's a lot easier to manage.
HCI provides efficiencies and mobility similar to that of the public cloud. We chose it a lot because we're still trying to do the private into the public cloud, that whole hybrid cloud transformation. We're still in the initial phases of getting out of our private cloud and starting to work some more clothes into the public. That whole process is very experimental
The footprint itself and the amount of man-hours needed to run the whole thing has reduced our TCO. It's definitely lessened a lot. We started cutting costs. HCI helps do that on the operational side.
It definitely shortens the time for us to deploy in production. We have a lot more lead time to plan, execute, and do all that stuff. It makes our business move faster.
Our hypervisor footprint has been reduced for both storage and compute. We started out in a data center with almost 60 racks of storage and compute, and we consolidated down to almost five racks. It's a nice improvement.
The Element software has enabled us to consolidate workflows and breakdown silos. Application performance has also been improved.
What is most valuable?
The ability to move the hardware around and the speed at which we're using all-flash are the most valuable aspects. All of our customers are very happy with the performance. We win on both sides of physically moving and the enhancements in the performance.
What needs improvement?
It doesn't have all the data points we need and all the historical data that I would like. I find that a lot of the performance analysis is done through support, where they have something that we don't have. It would be nice just to have all that on-prem. We haven't run into much else. It's pretty solid on the active IQ side and on the support side. I would like panes in everything.
What do I think about the stability of the solution?
We haven't had many issues with stability. We're still flexibly moving nodes left and right, so it's been stable to almost everything we need to do.
How are customer service and technical support?
I use support almost every week and so far, no problems. That's a big reason why we left whoever we left from before, was the customer support, and NetApp's been great in all aspects of that.
Which solution did I use previously and why did I switch?
We were previously using IBM but it felt like they were lacking functionality moving into the cloud. It felt like they moved very old school. They like to do things very 1980s and it was time for us to move on. We were definitely looking into a lot of automation cloud, other HCI platforms, but I think we sold on NetApp a lot because of their data fabric transformation, their hybrid cloud kind of story. That's exactly what we were looking for.
How was the initial setup?
The initial setup was pretty straightforward but we also had professional services and everybody there to help us deploy it. It was a pretty big deployment, plus it was kind of complex for us to be shuffling between three data centers. They helped us out and it was pretty painless.
What about the implementation team?
We have a partnership with a vendor who helps us and we also work closely with NetApp. Our partner helped us. They have a lot of expertise, but it was kind of new to them. It was nice having a partner knowing who else has used this, where's this deployed, who we can talk to, some kind of the market in business use cases that we are able to use them.
What was our ROI?
Our accounting people are happy. We're starting to dig into some of our budgets, so I think on most ends we have seen ROI.
Which other solutions did I evaluate?
We also evaluated Nutanix. We checked out Cisco HyperFlex. We took a good, in-depth look at a couple of other companies but I think we just ended up sticking with NetApp. We had a small NetApp footprint. Just FAS stuff and we decided to go all-in and go with HCI and the AFF line.
In general, they've always treated us well. They've always been great. I like working with anybody at NetApp. In general, we have a good relationship with both our partners and with NetApp.
What other advice do I have?
Dig into your workload and know your workload of what you're putting on there. HCI has worked great so far for just about everything but we did separate our performance-heavy, performance-sensitive EMR workload off to an AFF. We had to fix certain compliances with the application and HCI wasn't certified for it, so we couldn't run that on that platform.
I'd give it about a seven out of ten. It's not bad but we're still in our infancy, so we'll see. So far it's been pretty good.