Due to the fact that we did not have any kind of storage technology here at all when I first started out here, we purchased the SAN with three HPE ProLiant DL360 servers from Nimble and the solution we purchased was their dHCI implementation as it was supposed to do the setup and everything by scripts automatically. Right now, it's running my production and my DR environment.
I love InfoSite. That gets you a lot of good detailed information.
Generally, I use it for the DRF and I make sure that my VMs are all up to date. I make sure that they're all sitting on a host that's got decent resources. If they need to be moved, they'll get moved.
I like the memory management and I like the instant clone feature they have for the Horizons. That's what I'm using. I'm not utilizing the view, I'm utilizing their instant clone from VMware Horizons - vStorage motion, and things like that.
The scaling is great.
Technical support via the phone is helpful.
It's still kind of a new solution. I know when I purchased it, it really wasn't ready for prime time as I had to go through at least five or six deployments before it actually stuck. We were having a lot of issues. As far as where I'm at with it right now, I can't really complain. They're finally bringing it together where it can do updates through their one menu interface for the Nimble and that updates everything, the ESX hosts, the vCenter, it does all that for you. They've gotten better.
The only issue I'm really running into is we have a SQL server that we're having a lot of high IO latency on one of my drives and it's my data drive from my SQL server. It's not really giving me any good information as to where that latency is coming from, so that's what I'm trying to track down and troubleshoot. That's about the only problem right now.
The only thing I would like, and this may be more on the VMware side, however, going back to that SQL issue, I would like to be able to maybe have more granular control over my IO channels on my server. If I needed to troubleshoot why one server, one drive in a server, which is virtual, is having high latency, I should be able to go in and troubleshoot that specifically without having to go through and try to look at a ton of graphs. That's probably the only thing - just the granularity of it.
The reporting on Nimble to Nimble replication could be a little bit better
We've been using the solution for the last two years.
So far it's been pretty easy to scale.
In terms of usage, the software that the entire company uses is called Latitude. I run that on my Nimble storage. I've also got our development infrastructure on there. The virtual desktop infrastructure is running from there, Exchange, everything is running from the Nimble - my entire production environment essentially.
On the phone, Nimble support is really good when I call. That said, in terms of the guy that came onsite, I had to redo the cabling. He was a good guy, however, something got messed up and I had a lot of problems. I had to redo all the cabling. I'm not really complaining, it's up and running now. While they are great on the phone, in person the experience wasn't the best.
When I first started, they only had individual Hyper-V hosts and really no major storage. When we went from one to seven Hyper-V hosts that were individual nodes not even clustered, we decided we needed something that allowed for a cluster environment, and that's why we moved to Nimble.
We actually had to have HPE come on-site and set up the MLAG due to the fact that we bought two switches and they had to set up the Multiple Link Aggregation Group. That part of it's pretty complex. Other than that, with the setup now, if I purchased another server, plugged it in, and configured just the default that it needs, I can add that server pretty quick to my stack without any problems.
Originally, we had HPE come in and assist with the implementation.
For what it is, in terms of the scripts, if they worked initially the first time and the deployment was actually as easy as they stated it would have been out of the box, it wouldn't have been a problem. That wasn't the case. The pricing was a little high upfront for that reason, for me. Now that we've got everything running, it's not much of an issue. That said, we bought two complete Nimble setups with three ESX hosts, two M-Series switches, two UPS's, two racks, everything, one for here and one for my DR site. It was a lot of money we spent upfront.
We're just a customer and an end-user.
Right now, I haven't seen any issues. If a company wanted to purchase it, that would be something I would recommend as it does take out a lot of the guesswork as far as the networking, the storage considerations, the storage adapter of how it connects, and it takes all the guesswork out and sets it up for you.
The basic thing to consider is when you get it out of the box, you set it up, you turn it on, the scripts run, you have your vCenter, you just start spinning up VMs at that point. It saves a lot of time instead of doing all of that individually and setting up your own vCenter, the storage, and everything.
I was glad it had the scripts when they actually ran and it did connect everything in the backend. It did make it a lot easier. Out of the box, if purchased new now, would be a great option. Two years ago, it had a few problems and it's my understanding that those issues are mostly resolved.
I would rate the solution at an eight out of ten.
