What is our primary use case?
Prism Central is my daily management interface. There's a good overview of the individual clusters that we run.
We use it to take a look at IOPS for disk performance, disk latency, and storage statistics. We also look at the cluster memory resources to make sure that there's enough memory available. We also look at the CPU usage to make sure that there are not any unusual spikes or anything that could be played into creating a boot storm. If you're running virtual desktop stuff and they decide to recompose, suddenly there's a boot storm. We can get a feel for performance at any point in time.
We have four Nutanix Clusters in our environment, and the workloads go from database to general server and compute, applications, web services, and virtual desktop infrastructure. Our Citrix environment is run on a Nutanix Cluster as well.
It's all run in our data center.
How has it helped my organization?
We have greater visibility.
It also gives our infrastructure and our server teams better insight into, and controls over, how they stand up servers. Recently, Nutanix allowed templates to be built, similar to the way that VMware does things. Building servers, and eventually, being able to automate those tasks, is something that's going to make things a lot easier for those who build the servers.
And for the end-users who need access and take a look at the console, they can log in and they only have access to exactly their servers. It gives them a single pane of glass. They can see everything to do with their servers. They can customize their own dashboards for visibility into the particular virtual machines that they're responsible for. There's no worry about them clicking on the wrong server, consoling into the wrong box, or powering off or even deleting the wrong server. Those are all very positive things that the Prism interface has introduced to us.
Along with users having console access to those machines, it's great for us to be able to restrict certain users from hitting the console. There are some users that simply don't need to fiddle with it. They just need to see the current run or what kind of statistics their server is showing, or the uptime. They don't necessarily need to log in to it from the console. That's what SSH is for.
As far as security posture goes, we have put better controls around which servers our users can access and what they can do. We did that a little bit within our VMware environment, but not to the level of granularity that we have in Nutanix, with the roles and role mapping.
What is most valuable?
For me, the most valuable feature is that, when you log in, you have customized dashboards. You can add widgets to those dashboards and you can change the density of certain data points in the dashboards. When I log in, I quickly see where the cluster lies, as far as memory and CPU resources go. I can see how much storage we have left or if we need to consider purchasing an additional node to expand the cluster. That's what I first look at.
They populate the dashboards by default. Adding a widget is extremely easy. It's just a couple of clicks. And customizing the data density is a matter of choosing small, medium, or large. It's like: "What t-shirt size do you want to see?" It's a simplistic interface. Nutanix certainly takes its users, and the user experience, very much to heart.
There are also some great built-in dashboards that allow us to get some customer reporting performance statistics and to get a feel for what the future is. There's some reporting that does a "flash forward" to six months down the road with predictive usage for the cluster.
In addition, I have Life Cycle Management for updates. I'm primarily responsible for keeping the Nutanix infrastructure up to date with the latest code and the Life Cycle management takes care of that in a very small number of clicks. In other environments, I've run VMware with the Cisco offering for UCS, making sure that firmware and software all match up and that they don't fight. That was a pain. Nutanix's Life Cycle Management puts it all in one place and does all that legwork for me. When it comes time to run an upgrade, I click the button. It runs through the pre-install checklist, does a sanity check on itself to make sure that everything is inventoried properly, and then it applies the updates. I can walk away and have supper, or go to bed and wake up the next morning, and it's done.
We also have some disaster recovery and business continuity plans, using Nutanix and its data protection offerings. We have some protection domains that are set up to replicate virtual machines between sites. And that's something that was, once again, very intuitive and easy to set up. Scheduling those jobs and getting the status of those data protection jobs was very simplistic. If something fails, you get an alert email and it tells you exactly where the failure occurred.
For how long have I used the solution?
We made our initial Nutanix purchase in April 2020 and we've been using it since shortly after that.
What do I think about the stability of the solution?
I've had no issues with the stability of the system. All of the workloads are replicated between different nodes. If we have a hardware failure, the other node is there to take care of it, and then we can remediate by a reboot or hardware replacement. That's very simple. It is a very resilient solution.
What do I think about the scalability of the solution?
Scaling it is extremely simple. With the planning resource, you can complete a capacity runway report and it will give you a scenario for how many nodes you need to add and, potentially, what models to use. And to expand the cluster, you go into Prism, you click "Expand Cluster," and you find the new node and it brings it in. It's very simple.
How are customer service and support?
If there's ever a question, Nutanix support is always available. They are very prompt in their responses and very knowledgeable. We haven't had to worry about any information that has not been very clear to understand. They do a very good job explaining and keeping things on the level.
I've been involved in several sessions with Nutanix where we look at certain pieces of the interface, whether it be a task list, Life Cycle Management, or the list of your virtual machines. They have asked what data points we want to see for these machines. Do we want to see their IP addresses? Do we want to see the percentage utilized for memory or for disks? They put out calls and they take user feedback very seriously.
However, within the realm of hardware support, these are comments that I've already submitted all the way up to the top. With our previous storage solutions or hardware platforms, when there was a failing component, the replacement component was shipped out the next day. There was no user interaction needed for that process. With Nutanix, sometimes I have to open the ticket to begin the hardware issue, and there are some shipping acknowledgments required. The process itself is not extremely clear and I've been assured that that process will be improved in the very near future.
With any vendor, whenever there is a support issue, sometimes coming together and finding the root cause of that issue can take some time. And that's something that we've experienced with the interface. We were trying to enable multi-factor authentication and we encountered a bug in the Prism interface. It took a couple of months for us to locate the root cause and for them to be able to build a patch for it. We found a workaround, but it did take some time.
Beyond those things, we have really had a very positive experience getting Nutanix up off the ground and getting everything migrated over to it.
How would you rate customer service and support?
Which solution did I use previously and why did I switch?
Before we started using Nutanix, we used VMware which was our virtualization vendor of choice at the time. But seeing some of the offerings that Nutanix had, and our company's partnership with them, certainly made it a very easy choice.
How was the initial setup?
The installation process of Nutanix was very easy. Nutanix provided very knowledgeable installation engineers. The process was very simple. They asked for a lot of information in advance to make sure that we had all of the proper IP addressing and network configuration, and that our switches and everything else were prepared. So when it came to installation day, it was a couple of simple scripts and we sat back and waited for the cluster to update.
It was just me who was involved from our side in the setup.
When it came to migrating workloads, the virtual machines, and everything else, Nutanix does have a very nice tool, Nutanix Move, that helps facilitate all of those moves. We can schedule it and it synchronizes. But that can be time-consuming and it might not always be so intuitive for someone who is a little more basic. But it's a tool that gets the job done. You don't have to worry about restoring from backup.
I trained most of our staff on how to use Nutanix. We had about a one-hour session for most users. And for anything that they wanted beyond that, I referred them to Nutanix University, which provides free training. It's really a great site that I would suggest to anyone who is using Nutanix.
The solution requires software and firmware updates, but that's all handled through the Life Cycle Management that's built into Prism.
What was our ROI?
From what I understand, our maintenance cost for NetApp and the SMARTnet renewal cost for our Cisco UCS pretty much paid for our Nutanix environment within the first year and a half. We have certainly seen a return on our investment in maintenance and SMARTnet costs.
What's my experience with pricing, setup cost, and licensing?
The licensing structure is something that they're currently improving on. When we first purchased, we were informed that their licenses are transferable. If we were thinking of doing NC2, meaning Nutanix clusters on Azure or AWS, the licensing would be very quick and easy to swap over.
When we actually went to question them about that, we found that there were a couple of different licensing models, some consumption-based models and some licensing models that were tied to physical hardware. That was, perhaps, not the best experience. But the licensing structure is something that Nutanix, as an organization, is working on, and something that they're hoping to have improved in the very near future.
Which other solutions did I evaluate?
We didn't evaluate any other options for this particular move. The alternative for us was to buy an additional storage array and a different server infrastructure. We probably would have ended up going with VMware vSAN on Cisco UCS.
What other advice do I have?
We have evaluated running Nutanix in the cloud, that's something that's potentially on our roadmap, but we're also looking at just running servers in Azure, in addition.
I would certainly suggest that you request a test drive of Nutanix. They offer very simple test drives in their environment. Or get involved in the user group. Nutanix has a forum and research is key. Are there any lessons learned that I have personally learned in this process? Definitely consult with the network team before implementing, to make sure that there is proper capacity, but that's not a comment about Nutanix. That's a comment about infrastructure in general.
Which deployment model are you using for this solution?
On-premises
Disclosure: PeerSpot contacted the reviewer to collect the review and to validate authenticity. The reviewer was referred by the vendor, but the review is not subject to editing or approval by the vendor. The reviewer's company has a business relationship with this vendor other than being a customer: