What is our primary use case?
Our primary use case is healthcare for billing applications. With FlexPod, we use it mostly on some databases and billing applications. We are also using it now for containers, mostly with VMware.
We have the Cisco UCS M4 Blade Server, 6300 Series Fabric Interconnect, and NetApp AFF A800.
What is most valuable?
The number one feature is easily support. It is all converged. If it something breaks, it is easy to fix. It is easy to upgrade. These are some of the key reasons why we deployed it.
What needs improvement?
The majority of the time, if we need more storage, then we need to work with customizing the NetApp deployment. Right now, we just do a generic deployment, then wherever we have a need for storage, we have to move some application out of the next FlexPod deployment. One thing is to customize based on the requirements, but the requirements change so frequently, they are absolutely obsolete in six months.
I would like to see more artificial intelligence and machine language baked into the environment on the healthcare side. Right now, a lot of people are not leveraging AI, but we are in the insurance business and would like more flexibility by offering AI as a feature set into the healthcare environment.
What do I think about the stability of the solution?
The stability is pretty good.
The solution’s validated designs for major enterprise apps in our organization are very important. With upgrades and fixes, we can't afford downtime. That is number one. When you have multiple systems coming together there is always a chance of something not being compatible or something goes wrong. With this converge infrastructure, we know it has been tested by the companies. We know the issues beforehand, which is critical.
The firmware is all pretested and published. So, we do not have to go through the same process. That is how it impacts downtime.
What do I think about the scalability of the solution?
The scalability from the computer is pretty good. On the storage, they need to do something. They have to come up with some other options to scale both on the computer and also on the storage layer. An idea to fix this is possibly connecting the NetApp high availability model with a FlexPod by having them sit right next to each other.
How are customer service and technical support?
We have used technical support a few times. I'm mostly on the architecture side. The engineering team uses it. I hardly use the technical support, though I've used it in the past. It's good depending on the support level you get. We have enterprise level support. We have the highest level support from Cisco and have never had an issue.
The solution’s unified support for the entire stack is critically important because we cannot afford downtime.
Which solution did I use previously and why did I switch?
Before, it was on a Cisco UCS C240 M5 Rack Server, and we moved some of the applications on a very limited use case. With the innovation of the AFF A800, its ease of management, and supportability, we have seen some performance improvement with the solution. The performance has improved by two or three times.
How was the initial setup?
The setup was fairly complex because of the sheer number of servers, more than 30,000 servers.
Once deployed, it is set up and forget it. We do not have a dedicated FTE to manage this solution all day long. That's a good thing.
What about the implementation team?
I highly recommend if you're deploying this, do not deploy this on your own. Definitely work with partners. That is my number one recommendation.
What was our ROI?
We don't have a TCO model right now.
Which other solutions did I evaluate?
We have a very strong Cisco partnership. All our networking stack and some of security stack is all Cisco.
VxBlock was also on our shortlist.
We chose FlexPod because we already had NetApp deployment onsite (on-prem).
The history of this product's innovations affects private hybrid cloud, mostly. We have a VMware cloud foundation running on FlexPod and want to take this to the next level, either VxRail or on HyperFlex. Those are the solutions that we are looking at right now. I think they are working on SEEBURGER as the next step, but maybe we might introduce NetApp HCI.
What other advice do I have?
I would rate the solution as an eight out of 10. I would suggest or recommend FlexPod for deployment if you are moving from a predefined converged infrastructure or validated design architecture. Though, you have to customize it based on your requirements. Right now, do not just jump in. Work with a partner to build out your requirements, then deploy it properly.
Our data center is huge, so it has let us reduce some cost, but nothing significant.
Which deployment model are you using for this solution?