NetApp FAS Series Reviews

The integration with VMware is the most valuable feature for us because we run a lot of VMs and the backup is very good when you run your VM in NFS.

We had a case when they had to restore a lot of data. We went back one hour and got back everything. The restore itself only took about an hour.

Some of the tools could be improved like NetApp OnCommand. This has been a lot better recently, but they could make it faster.

We've been using it since 2005.

It's very stable and I’ve never experienced any problems in 10 years.

It scales to our needs.

8/10 - only because it is impossible to have a 10, as there is no one that good. We’ve had a good experience with their customer service.

The solutions that are present on NetApp’s website are enough usually, but when it is tough for me to resolve it on my own I go to our consultant.

We did a long time ago.

Initial setup was pretty straightforward. We started on a small scale and built it up.

We implemented it in-house, but we use a consulting company to help. Now, we run it on our own.

It fulfills the needs we have for storing data well. We had a lot of storage spread out over many devices from many vendors and now everything is consolidated. It saves a lot of time.

Ask other people who use it as references are really valuable.

• High Availability – most important thing with all big storage systems
• Flexible – it uses all different kinds of protocols (SAN, fiber channels)
• Virtual Management Console – helps with and integrates with vSphere. I’ve deployed it, but haven’t configured it. Looks really promising.
• Snap Drive – integration with Windows for SQL exchange is critical
• Can update without taking the system down

It’s a High Availability environment in which I haven’t had problems. We’ve only had one disk fail in a year. Quality of product is good.

Tegile offers live deduplication. And NetApp can only be scheduled to dedupe in the background or for later.

Also, the web page for downloading software could be more efficient (for example, getting instructions takes a few clicks). Make it easier for customers to download software.

The instructions didn’t provide for re-configurations from scratch and tech support were stuck too. The person who configured before had installed NetApp 8.2 on 7 mode. 8.3 didn’t support 7, which I wanted to do, so that’s why I had to reconfigure from scratch.

First, make a lot of calculations of how much data you’re going to use and how much you’re using now. Compare the two and look into how you configure storage solution. How much data, how to split, etc. If you can’t spend that much on storage, you need to determine how to follow best practices to get as much storage as possible for the money. For example, if you have five shelves, you don’t need to worry about LUNs going offline. But if it’s just two shelves, like I have, you must be very careful about provisioning storage and use as much as possible.

The performance allows me to provide backend storage for large number of VMs and databases at a competitive price point.

Unified Monitoring v6.2 loses a bunch of functionality that previous versions had. For example, I took a cluster out of Unified Monitor, but Storage Monitor was still alerting me about it. 6.2 is not as comprehensive, but Unified Monitoring 6.2 will only be useful when it does everything. Insight’s price is just too expensive and unreasonable.

It's pretty stable, even if it runs into something freaky, it keeps going. For example, mysterious a reboot, and nobody notices. It keeps working.

It scales to a point, and then you buy more hardware. Doing a head swap (swapping out controllers) is not as easy as it used to be.

It's better than Oracle, but actually pretty good. They're responsive, and help resolve situations. We have had a couple of issues, but 99% of the time, they get me an answer, although it may not be what I like, but it’s a definite answer within a reasonable time frame.

It's complex, not a trivial task. We can unbox it and deploy. There are many unpublished tech tips that NetApp engineers get that customers don’t (for example, how to save a disk).

The price-per-gig makes it the most expensive storage, more than EMC VMAX. So I’d like to see more aggressive pricing.

It's losing points on its value. The performance is nearly perfect, but it’s really expensive.

• Clone – we updated to Dynamix AX 2012, and not a week has gone by that we haven’t fired up a clone to test a user, code, etc.
• Disaster Recovery suite of functions
• Quick backups that don’t impact system
• Refreshes test environments quickly

• Reliability – hardware performs well, we’ve never had problems
• Great support from Sirius
• We’re able to recover a file if we need to

We use mirroring a lot, and if it had snap manager for SQL included, we could do that from one location.

Rock solid. I’ve been at Wilson for three years and have also used it before at another company. We have a disk go out every once in a while, but no issue with FAS.

In the last growth, we went from a 3250 to an 8020. We’ve set it up to grow out easily by just adding trays.

Lately, a little slow to respond, but once you get someone, they’re very knowledgeable, although I haven’t had to use it a lot.

I was familiar with 3250’s from another company, so I knew how to use the software, but Sirius helped us. So in that regard, it was straightforward.

• Software features, such as being able to do snapshots and file system optimization
• High Availability -- components fail so this is a nice feature to have when failing over. There's no downtime, so we don’t lose data.

Good bang for the buck. Also, we use NFS generally, but FAS has the capability to use SAN, so it has a broad spectrum of use.

Tough for me to answer because I’m limited in my role, but the one thing I’d like to see most is more cohesiveness with a unified manager. I like the end product, but it’s not really all integrated and is convoluted with different managers. I would ike a single pane of glass, a single dashboard.

We see a lot of bugs in roll outs, and sometimes I think the first GA are late-beta deployments. My impression is they could have let it bake a little longer. But it could also be because of some of the environments it deploys in.

Snap Manager v3.3.1 is a little buggy and NetApp doesn’t offer training course on it. So it could be what I’ve been taught by other people, or it’s in fact buggy, but likely a little of both. Hopefully they made improvements on 3.4.

7-mode scales very well. I’m even more impressed with where they intend to go with cDOT, but it may be rolled out prematurely.

Tech support is usually pretty good, but occasionally there are some things that occur only on our site that tech support has issues.

Plan ahead and make sure you right-size it. How much head room do you really need? How many spindles are you going to attach? Are you really going to share workloads or do you want to separate some of those? We don’t segregate our infrastructure, which I don’t like, but all that costs money. But you should make sure that you have failover.

Snapshot, because so much of it is on our end-user storage, our users often delete things they’re not supposed to. Having snapshots to revert these deletes quickly and easily is very valuable.

Our greatest advantage with it is ease of use, flexibility, and reliability.

Knowing what’s coming down the pipe, NetApp is headed in the right direction. In their five year roadmap, it provides what I need it to do.

It's extraordinarily stable. We had one outage one-and-a-half years ago when batteries were bad, but that was a known defect on that particular model. However, that was our fault for knowing this was an issue. We've had two outages in 10 years due to something other than operator’s error.

Incredibly scalable. Not even touching what it could do. Between scale up and scale out, we’re not even close to reaching its highest potential. We have a four node NAS with the potential for 24 nodes.

It's fantastic.

Once you’ve done one, it seems very intuitive. However, the first time seems very complicated.

Of all storage technologies I work on, it’s the easiest to learn and one of the most powerful. But you need to spend your time taking classes before digging in too deep. Get educated.

NAS functions, as it's primarily used for all our file shares. We have other NAS devices, but this is easier.

Also, High Availability is a valuable feature.

Snapshots are good, especially the snap mirror, which we use for disaster recovery and backups. Also, we have a lot of data centers (seven primary centers) and we deploy at each of them.

I miss their old support structure. We used to be able to call up and get an answer pretty quickly, but now it’s more arduous.

It could be cleaner for dedupe, and I wish we could do dedupe for the entire system and not just a specific volume.

It's highly reliable, but has had the occasional bug. We install patches or shut off features.

Depends on how you’re scaling. If wide, it works well. Vertical scaling not so well because we’re primarily SMB. No matter how brief, people don’t like being offline (e.g. baby monitors).

I’ve worked with them for over 10 years. They used to be stellar, but in the last three to five years, not as reliable. The quality of information you get from them is less specialist, and they've not broken it up so that you get routed to a particular technology, it used to be one senior guy who knew everything.

There’s always networking issues, but not related to NetApp.

Other than tech support, it loses points because it could always be better.

It depends on what you’re implementing. Consider carefully what you want to do – for example, have enough vLANs because you don’t want to be adding more later.

• Dedupe
• Also, our customers look for fast connectivity and cost efficiency.
• It's TCP/IP vs. fiber channel, which tends to be more costly.

• Single unified-type architecture for block-and-file-type data storage
• Ease of use
• Being able to hand off things like snap shots directly to customers

They need to improve the go-to-market for all-flash and converged infrastructure. What is your goal-to-market vision, and when to get there? They’re too slow compared to others and what they’ve done in the past. They were the leader in dedupe, but now, it’s not such an innovative edge.

It lacks flexibility in failover and failback, so we cannot granularly failover pieces. It's not easy to move one piece over to the other side.

Also, from the overall workload standpoint, all protocols are handled in just one physical architecture. So if I'm running dedupe, fiber channel, and other protocols on the same CPU core, I can’t load-balance. I’ve seen issues specifically with EMP, one core is maxed out, and I can’t use the other cores to handle it.

Fairly solid 5-9 array. FAS is a solid architecture in 90% of the environments.

Scalability especially in SMB range has been well-received. So long as the environment is sized correctly, it’s been good.

I have had both good and bad experiences, depending on what tier I get to initially. Now it’s tiered, whereas it used to be one senior guy.

If historically you’re a NetApp customer, it’s not as complex as cluster mode. It requires a lot more complexity – command line is not so friendly for storage admins. I’d recommend also sticking with what you know.