FlexPod XCS Reviews

The ease of use. I like that with all the NetApp products, it's very easy to use. I'm on the storage side of it, but we're still working with the servers and we're using the Cis UCS servers. So there are some tweaks that they're still working on on that side of it.

But the original PoC we did came back with very good numbers and looked like it was going to help a lot of our users locally. And we have remote users that will attach to the FlexPod and do their work from say, Houston, logging into Calgary or other places from outside of Calgary. It centralizes everything into our datacenter and it helps us manage it more easily.

The general manageability of it really is easy and it's taking advantage of all the deduplication, compression. We've got 9.1 P3 in there now so we haven't taken the next step to do compaction or anything like that but I'm sure we're going to go down that road too.

The benefit is the speed, it's the performance. That's what it comes down to is the performance of the solution.

It's the first step of it right now for us. They're still proving some of the server side out from it but we've already pushed out the storage side of it. They're using that storage in their existing, and they are getting better performance, better benefits.

I'm not really sure, to be honest. A lot of what it's doing today is exactly what we need, so I'm focusing on some other things at this point around our database environment and things like that. Everything that I've seen right now from a FlexPod perspective is very good.

I think they can always improve, whether it's dedupe or compression, those algorithms; and flash through better SSDs. I guess faster is not really there with an SSD now, but I think anything that makes it smaller, better cooling, less power, those kinds of things. Help in the datacenter.

We've had FlexPod in our environment for, I would say about six to nine months. We brought it in to help one of our applications called Petrel. It's IO intensive. It's an application that petrophysicists use to look for oil. That's what they're using it for, so they need performance for projects.

In general, just because I've been working with NetApp products for a lot of years, from that perspective, I don't have outages that I worry about. It's very stable in that sense. I have more problems if my network goes down than anything else, and it's not my problem anymore.

Very scalable. We're already planning how we're going to expand and grow it. They're doing a lot of exploration work so we know it's going to expand so we're already planning for it, and it's going to be easy to do.

We've used lots of technical support. I've used it. We've talked to people on the phone, we've done the chats online. All kinds of different things. NetApp's support's always been great for us. They're knowledgeable, absolutely.

I was not involved directly in the initial setup. The other guy that I work with was part of the PoC that we did, and then I got a little bit involved later on. He's actually part of the finishing off of it. But the ongoing support and operational part of it, I'm part of that.

They had to jump through a lot of hoops to get things but I think part of it was our own... What we needed to do was listen to the experts a little more to say, "Okay, we have to do it this way versus trying to push our own infrastructure on it." So that's what hurt us. But in the end, really good.

FlexPod is absolutely uniquely valuable for oil and gas, for some of the more higher-intensive products, the software that engineers will use to look for oil They're always looking to be able to do that faster, better and more efficiently.

I gave it a nine out of 10 because nothing's ever perfect. We did run into some hiccups around some different things. Part of that is us, the other part of that is working with the vendors. We weren't utilizing things with the switch properly. When we were having jobs come in from outside to access the storage, because it wasn't going through the Cisco switch and everything else, it was actually a slower grab for them. Once we changed those things and really implemented more of what you should be doing with the FlexPod, that's when we started seeing the performance gains.

That it works. That it does exactly what it says on the tin. That once it's set up, it does exactly what's its supposed to do. There are no "gotchas," there are no "oopsies." Not in this particular use case. There is no hidden BS that has to be satisfied, this, that, or the other. It just does exactly what it's supposed to do.

Up time, cost, predictability.

With the old Dell EqualLogic, EMC, up-time was always an issue. There was always something that created a problem here and there. We have, in four years, not had a single system-down issue. That's hard to beat.

The predictability of our system utilization: I can predict fairly easily when I need to go out and buy new shelves because I can trend it all. I can fairly easily predict where I need to open more aisles. I can fairly easily predict where I need more space. It just works.

The CLI part of it is still evolving enough that commands that you expect to do something become deprecated and you want to take their place and you have to keep up with the code base. In this code base you do this, and in this code base you do this, and in this code base you do that. But for the most part its good and, let's be honest, everything changes.

Automatic tiering would be good to have.

My biggest thing is I would love to see a native SMB or NFS front end to an optic store on the AFF and FAS platforms. Right now you want me to go out and buy a front end for it that creates an optic store on it and gives me the SMB interface. I would love to see that as a native part of the SBM. It doesn't have to be the end all be all; it doesn't have to be this hyper-scale thing but just the fact that I have it, so I can dip my toe in it, and I can get something that kind of works, that would be epic; that is my main thing.

The other big annoyance I have with Net App is the fragmentation of all the software. I have SnapManager, I have this, I have that, and they're all slightly different. They all look slightly different. They all come in different VMs. Some are OBAs, some are installed on Windows, some have weird requirements like, "No, no, no, it has to run on this version of Windows." It would be nice to just have all of that in one giant application and then just turn on and off different features based on license keys. That would make things easier.

About four years.

Great. It's been perfect for four years. You can't beat that.

So far I have not had any issues at all. The only couple of things I would like to see would be something, as I said, like internal tiering where you could automatically set up an aggregate spread across a 10K disc and have the controller automatically tier it. But now we're going all flash anyway so who cares? We've kind of brute forced our way out of the problems.

They're good. They're knowledgeable, absolutely. I have no complaints with the tech support that we have had to deal with.

We switched because EMC and Dell, and EqualLogic sucked and it was driving me nuts and it never worked.

It was a paradigm shift because every stack has its own set of unique ways of doing things and getting used to that and getting into that mindset took a little bit of effort but once you get it, it's clear sailing.

Upgrades are not complex.

We're a law firm. I think this product is valuable for pretty much anybody who has a large amount of data that they need to manage. I don't think that this product is uniquely valuable for a law firm.

I am a very happy customer.

• Pretty much a single phone number to call when there is a problem.
• Ease of use
• Manageability for the storage system

It is valuable, because it is easy. Easy to use, so we don't spend a whole lot of time learning new products out there. That is a major plus.

It definitely made an impact to how we manage our data. We recently have a new data center, and we migrated our data from one of the older storage solutions out there to an all NetApp environment.

Tighter integration with CISCO.

Since 2013.

It's very stable. We haven't really had any major issues with the FlexPod solution in all four years.

I find it very easy to scale upwards and horizontally, as well. It's very easy for us to scale up by adding either additional controllers or additional storage shelves.

We have used tech support. They are very knowledgeable and very prompt in getting back to us.

We have engaged technical support to assist us on a lot of different things. Whenever we have a case open, and if we think it requires some escalation, we have a SAM and they make sure we get the resources that we need and get back to normal operations within a short period of time.

Before Flexpod, we had other Meta app systems going back to 2008.

We're a health system company, so we have a number of different storage solutions. However FlexPod, it has everything.

We still have a few solutions because some applications have their specific storage systems, and being in the health industry, those applications are usually approved by FDA, and it's not something which can be changed at will.

I was involved in the initial setup. It's pretty straightforward.

It simplified our server farm. We were able to consolidate down to one rack from the three or four server racks we had before. We were running a lot of servers, which are supervisory control and data acquisition systems for power systems. We also run a lot with OSIsoft's PI solution.

It definitely gave us a more robust system than our original, old, individual servers. It also simplified management, both on the network side and the server side. It saved us a lot of time. It probably cut our management time into server-network troubleshooting, or just normal management, by 40%.

Right now I can't think of any, because we're so specialized in our environment. I think maybe going to a full solid-state would be beneficial. I don't know how beneficial it would be for us in the power industry, because a lot of our equipment in the field is maybe 20-30 years old.

We are interfacing with a lot of older devices. We're using the Fabric Interconnect back to our Nexus chassis, so I don't know if we can go up to 40GB yet. It's probably just added more speed, but we're limited by our connections out to the field anyway.

It's been very stable. We had one or two issues with a spinning disk, but there has been no impact to the network as a whole.

Scalability has also been great. We have been able to spin up new virtual machines as needed. We haven't run into any bottlenecks.

My server side technical lead worked with technical support more than I did. I think he's been very impressed with how good and responsive they are.

We needed to migrate away from our older servers. When we did the cost analysis through the FlexPod, and replacing each individual server, it just made more financial sense with the FlexPod in the long term. We were using individual Dell servers, or HP servers, it was kind of a mishmash.

I was involved with the initial setup as the network engineer at the time. That went very smoothly. The most surprising thing was when I connected Cisco Prime and had it search for a new sister device, it pulled in that app, or the UCS part of it, the fabric, and the connects, automatically.

We had looked at the EMC VNX series. I wasn't too involved in that, I only got pulled in when it came in to interfacing with the network.

We chose Flexpod over EMC because FlexPod had the Cisco commonality to it. That was one of the major reasons we went with the FlexPod. I had some experience at a previous job with the VNX, and that was a very good solution as well. But, for our environment we were trying to standardize on Cisco, and that was a big selling point.

I'd say, go for the NetApp with the Cisco UCS. It definitely will cut your management time down and it's a very reliable solution.

The most valuable feature is the converge nature of having compute storage on the network in one rack. We are able to use the virtual switching to add multiple networks, DMZs and sub-nets to be able to use the resource of the FlexPod itself. We are running the full gamut for minding business applications and web applications. We're also using the virtual switch pieces to run our Citrix environment that's also running on FlexPod.

It's greatly reduced the complexity of network storage and compute. It allows us to deliver services quicker. We can provision faster, we have great use of templates, so we can spin things up in minutes and provide services.

It probably already is in the product now, but at the time, we didn't have a really great SSD shelf that you can just plug and play in there. I know it is present today, but that was the only feature we were looking for at the time. There are probably some nuances in how the network and those types of things work. Maybe it could have more templates. Other than that, it's a great product.

We've been using for eight to ten years and we never had any downtime. It's been a great product for us.

In terms of stability, FlexPod has been great. I don't think we've ever had any downtime.

Scalability's been great. We've actually added shelves, we've got blades, and we've added storage. We even extended our presence inside the environment because of what we like inside the FlexPod itself.

In terms of technical support, it’s all been great. We have Phone Home support, which we haven't used very often at all. Any time we've needed any support at all, it's always been first rate.

I was involved in the initial discussions, evaluation, and purchase decision. My IT team was actually the hands-on team that did the setup of the FlexPod. Once we got it installed, got a little bit of knowledge transfer, they really enjoyed it. The solution brought everything together. It made it a lot easier to deliver services, because compute stores and networks weren't these discrete components they had to work on separately.

There wasn't that much hyper-converged at the time we went into FlexPod. It really was the game changer at the time.

We were looking at a new VM farm to start with, so we were looking at Blade Server solutions from HPE/IBM. What really did it was that FlexPod had all the components in one rack that we could basically turn on and forget.

We continue to evaluate solution on the market. We enjoy the UCS environment and that's the way we increase the blade. We like the way it integrates with the NetApp, so the FlexPod just really brought it all home.

Take a look at the product. If you go through your requirements and have to redo either your virtual environment or some of your SAN storage, look at putting your compute storage and network together.

Number one is the ease of deployment and number two is reliability of the solution. We are running electronic medical records applications and also document scanning, in addition to standard infrastructure type servers.

We are using Epic, OnBase, 3M, Nuance Transcription, and then infrastructure servers, BNS, Active Directory, SharePoint, and SQL Server. The performance is excellent.

Benefits for your organization are probably the reliability and the cost. We feel like it's a very cost-effective product, and the fact that it doesn't take a lot of people to support. We're very thinly staffed, so just those two things are really reliability and then cost.

I would like to see more scalability and possible cost reductions with the all-flash solution. I would like to see a focus on database optimization on the Intel platform. This could be with Windows, or HANA, or even cache, down the line, something with the Linux cache. I would like to get something that is optimized for those databases.

We have two camps now. We have to have a AIX camp and then we have an Intel camp. If we could just have it on the Intel platform, have FlexPod take it all, it would be easier to support.

We have been using this going on four years.

It is a very stable platform. I've had minimal downtime in the entire time we've used it.

Scalability has been very good. There really has not been an application, with the exception of the Epic Cache Database, that we haven't been able to use with this solution. Anything that we run on the Intel platform, on the UCS platform, we've been able to use in order to scale.

Technical support is excellent. They've been really good to work with. I think they work well with Cisco. We've never gone into like a finger pointing situation at all. We also use Cisco ONE, in which they'll take on the whole stack. It is very cohesive. We haven't always used that and we just started using it. That's an even better solution, in that it is only one call. Before, we would make two calls: One to NetApp and one to Cisco, and they would then work together. It wasn't a problem, but this just makes it easier from an escalation standpoint.

I was involved in the initial selection process, purchasing, setup, and the whole thing. The setup was very simplistic, which is one of the reasons we went for it. We try to strive for simplicity in our shop because of our staffing situation and because we are non-profit organization. So reliability, efficiency, all those things really come into play. We're a 7/24/365 facility, so things just have to work.

We evaluated EMC, Pure Storage, Tintri, and Nimble. We decided to go with NetApp over the competitors because we did an executive briefing with them in San Jose.

We were really impressed with their road map and their direction. We developed more than just a vendor-customer relationship. It was more of a partnership. We felt we had a good relationship and that we could trust NetApp, and that the solution would work. It was really, in the end, the technology, the price, and the people.

I would definitely recommend FlexPod. The first thing I would say is just the simplicity and the standardization that you can put in place. You can do more with less. This packaged solution already works, so you don't have to figure it out and be your own integrator. This is the way to go.

It has simplified our server farm. We were able to consolidate down to one rack from the three or four server racks we had before. We were running a lot of SCADA servers, which is a supervisory control and data acquisition system for power systems. We also run a lot with OSIsoft's PI solution.

It definitely gave us a more robust system than our original, old, individual servers. It also simplified management, both on the network side and on the server side. It saved us a lot of time. It probably cut our management time of server-network troubleshooting, or just normal management, by 40%.

I can't think of any improvements, because we're so specialized in our environment. I think maybe going to a full solid state would be beneficial. I don't know how beneficial it would be for us in the power industry, because a lot of our equipment in the field is maybe 20-30 years old.

We're interfacing with a lot of older devices. We're using the Fabric Interconnect back to our Nexus chassis, so I don't know if we can go up to 40GB. It's probably just having more speed, but we're limited by our connections out to the field anyway. Speed would be the area where we would like to see room for improvement.

It's been very stable. We've had one or two issues with a spinning disk, but there was no impact to the network as a whole.

Scalability has also been great. We have been able to spin up new virtual machines as needed. We haven't run into any bottlenecks.

We have used the technical support. I know my server-side technical lead has done it more than I have. I haven’t heard him complaining. I think he's been very impressed and the responsiveness has been very good.

We needed to migrate away from our older servers. When we did the cost analysis through the FlexPod, and the cost of replacing each individual server, it just made more financial sense going with FlexPod in the long term. Previously to this solution, we were using individual Dell and HP servers. It was kind of a mishmash.

I was involved in the initial setup. I was the network engineer at the time. That went very smoothly. The most surprising thing, was when I connected Cisco Prime and I had it search for a new Cisco device, it pulled in that app, the UCS part, the fabric, and the connects, automatically.

We had looked at the EMC VNX series at the time. And that point, I wasn't too involved. I only got pulled in when it came to interfacing it with the network. They chose FlexPod over EMC, due to the Cisco commonality to it. That was one of the major reasons why we went with the FlexPod. We knew Cisco, and we worked with Cisco already. I had some experience at a previous job with the VNX, and that was a very good solution as well. But, for our environment, we were trying to standardize on Cisco, and that was a big selling point.

Go for the solution with the Cisco UCS. It definitely will cut your management time down, and it's a very reliable solution.

The most valuable feature is the defined architecture, so you know you're going to be using best practices. That's key and important to us. We use the segmented layer 2 architecture that we got from a design on FlexPod, and that's helped a lot.

As opposed to just applying something like a NAS with systems, I don't know that this product necessarily does improve the way I work. We haven't changed a lot of our practices. We follow best practices, generally, anyway. We haven't engaged; we haven't had an issue where we've needed to engage Cisco and NetApp together. I think that would be a benefit if we had an issue or we needed to get everybody involved.

I’d like to see cloud features, for sure, and auto-scaling type things would be good. Automation is important, and that will be more important going forward.

I don't know what they would need to do to earn a better rating from me. I know that, when we do block workloads on NetApp, fiber channels specifically, it hits the filer pretty heavy. I don't know why that is. We're going away from fiber channel anyway.

It’s stable.

It is very flexible and very scalable. We've grown our cluster, so we haven't had a problem.

For tier 1 cases and system down type of issues, NetApp support is very good; for OFFTAP, not good. I've provided that information to them many times. Their OFFTAP support is not the best.

We were doing a technical refresh of our systems and of our storage. That's where we combined that to use a FlexPod-type architecture, to do both of those things and make sure that we're aligned with best practices.

Determining the architecture was complex, because you need to make sure that you know what your requirements are. Then, once you've designed it, initial setup was straightforward.

We looked at other vendors. We keep our eyes open all the time.

The replication was why we chose NetApp, SnapVault. Not a lot of storage vendors do block-based replication with being able to maintain a different set of snapshots on the secondary and the primary. Everybody does SnapMirror, or does a mirror of some type, but SnapMirror XDP or SnapVault is something that NetApp has that most storage vendors don't have.

In general, global- and enterprise-level support is the most important criteria when I’m looking to work with a vendor.

Spend the time up front to architect it, get the details, and make sure your plan is solid.