NetApp ONTAP Review

Has high availability and enables us to centralize mission-critical applications on the MetroClusters


What is our primary use case?

We use pretty much every part of the product in our company. The core of our company is in the United States and Germany and we are using MetroClusters usually for that kind of workload.

How has it helped my organization?

The product helped improve the way our organization functions because we are able to centralize mission-critical applications on the MetroClusters. Previously they were spread out in our environment and now it is easier to use and administer them.

What is most valuable?

The high availability is one of the most valuable features. But it is also nice that it is easy to do snapshots and to recover things using SnapVault. Those are great features.

What needs improvement?

I think everything can be improved, really. The IT world is evolving all the time, so there's always a new challenge and always a new thing to improve. It is harder to say what that could be. In a perfect world, there should exist some kind of guide as to what kind of workloads you would be able to mix within the same cluster or aggregate. That would be something really good and useful for us because we have centralized everything in just one MetroCluster. Now we are not sure if that architecture is actually giving us a few headaches or if it is something different. We have more than 15 petabytes of space use and we have a really low percentage of failures. In the boxes that are not really I/O (Input / Output) intensive, they are working very well.

What do I think about the stability of the solution?

The standard allowance for HA (High Availability) is really good. We do not have many usage problems with that. For the MetroClusters, we have had some problems with latency related to some calls on the backend. But besides that, it usually is a very stable environment in the MetroClusters.

What do I think about the scalability of the solution?

The scalability is good. We started with a two-node MetroCluster, then a four-node MetroCluster and then an eight-node MetroCluster. The eight-node MetroCluster is giving us a lot of headaches. The four-node MetroCluster — if it's correctly sized and configured — it is really good.

How are customer service and technical support?

Usually, technical support is really good. Most of the time they have a really good response. I really like the Zoom sessions because we can solve everything right away. That has been a great improvement from their support services.

We do feel that they have been supportive with the bigger issue that we had. In this case, we understand that a lot of people are involved. The thing is that I think there are some technical challenges for support and other cases being driven by the engineering level. So I know it is a difficult case and eventually, they will resolve it.

Which solution did I use previously and why did I switch?

We never needed to invest in a new solution because we were always using NetApp. We were using Windows Servers before for NAS (Network-attached Storage) environments.

How was the initial setup?

The initial setup was very simple. It is just making the purchase and getting up and running with a technician. We already have WFA (Workflow Automation) workflows for that and to deploy our standards, so it is quite easy. In just two or three days it was done.

Which other solutions did I evaluate?

We have to run proper procedures for discovery, so we invited every vendor in the market and then we continued working with them until we narrowed down the solution to only NetApp. The reason we ended up choosing NetApp is that it has great features other products don't have all of together. Features like Snapshot. 

The support is really good also. Usually, the releases of firmware are not that good because we usually get a lot of bugs that we need to address. We like the hybrid models and the all-flash modes. We have all kinds of sizes and flavors, but those are the ones that really work. We haven't implemented anything for block-level yet, but maybe in the future, that may be on our roadmap.

What other advice do I have?

The solution's Snapshot copies and VIN clones work very well and they are easy to use for recovery. We have a huge environment. We are running around 15 petabytes of data, so doing backups and restores is a daily job and these features have done marvelously for our environment. It speeds things up and it is really easy for us to manage, especially in that size environment.

Snapshot copies and VIN clones also affected our application development speed and made it way easier. We are not using containers so that has sped things up a lot. We love that feature, really.

Consistency of storage management affected our storage operations by helping to reduce our organization's data footprint. All the space savings — on-prem, of course, because we do not use the cloud — is honestly really good. Overall, we experienced around 30% savings.

Apparently we have a lot of latency but it is not related to the protocol. It is something within the operating system of data on top. It is still a question we have open with support and we have no answer so we do not know what the problem is yet.

On a scale of one to ten where one is the worst and ten is the best, I would rate this product as an eight. The eight is only because of the problems that we already had. We would expect that a MetroCluster would be able to provide huge performance and we had so far three outages during this year. That is quite a lot really. Overall I think it is still a really good environment and using NetApp has been a really good solution for us. But outages hurt our experience overall. It would be a ten for us if we had 100% uptime.

The advice I would give to people considering this solution is to be very careful about the kind of workloads they are running. Keep a really close eye on things. It is best to use all the features of NetApp rather than just buying isolated boxes.

Which deployment model are you using for this solution?

On-premises
**Disclosure: I am a real user, and this review is based on my own experience and opinions.
Add a Comment
Guest