FlexPod Review

Our downtime has significantly been reduced

What is our primary use case?

We have four different use cases that we bought it for:

  1. Our production VM or infrastructure is on a FlexPod with a metro cluster.
  2. We have a CCTV system, which is a FlexPod using E-Series as a back end.
  3. We have another E-Series FlexPod for backup infrastructure, with our combo products.
  4. We have a test end environment, which is a mini replica of our production, VMware assistance.

How has it helped my organization?

We have been using FlexPod for seven to eight years now. It has evolved a lot over time, primarily in ease of connectivity. It has been built around all the same platforms. It is just what storage back-end that we decide to tie it into it. Will we be using blades, a chassis server, or rack mount servers? This makes it easy for us, because everything is consistent. 

It does not matter whether I bought one five years ago or if I bought it today. All of my connectivity will be the same. When I put it in the data center, it takes a few hours, then I can have a base system up and going.

What is most valuable?

I work at a state agency. With FlexPod, I can contact to NetApp. I can contact our rep and I can get the building materials from him which includes all of my switching, servers, and storage in one place. It saves me a lot of time when I have to go out and send out a bid, especially the bids for larger dollar amounts and longer terms. The more efficiently I can get those bids out and processed, the better it is, and the faster I can deliver solutions to our customers (our users).

What needs improvement?

There are a few nuances. There is always something which bug you. It always seems like we run into the bugs. It is usually just a simple code update or something like that. 

There is always room for little tweaks and little improvements to make life easier. A few things, the E-Series is stupidly, simple. However, the FAS in it, with all its flexibility and scalability, it is much more complex and could be simplified.

We had not upgraded to the most recent release of ONTAP (and some of the other newer tools). The newer version that we are in right now went from an Clustered ONTAP 8.2 to an 8.3. In the 8.3, some of the stuff disappeared. It is there, but it is not intuitive to navigate to, like the IO Statistics, etc. I hear this will be fixed in the next versions, but we have yet to see it. 

It would be great to see some form of interoperability between the FAS units and the E-Series, specifically for replication, even if it is just one more replication from a FAS to an E-Series. That would be amazing. 

For how long have I used the solution?

More than five years.

What do I think about the stability of the solution?

Considering we are on our second generation of FlexPods. We are fairly happy with them. For the most part, the system is a rock. Whenever we have needed it, it has always been there. That is the key

It has never fallen out from under us when we were trying to do a critical push.

What do I think about the scalability of the solution?

We haven't hit any limits as of yet. Our most recent purchase is actually a 3.6 petabyte raw system. It is a 360, 10TB drive, which, at the time, is the largest they could do. It is an E-Series, it is more storage then we know what to do with right now.

We are only using about half of it right now, so the scale out to the future allows us to get where we want to go. We use it for a CCTV solution, so video never gets smaller, it only gets bigger because there are more cameras. More cameras with a higher resolution and higher frame rates. We made sure that we purchased a system which would will grow with us and scale with us as we need it to.

How are customer service and technical support?

Unfortunately, we had had to use technical support a few times, but for nothing major. We have not had any major failures, usually it is just your typical drive. We have 1500 spinning disks, so we have a drive die here and there. Most of the time, we do not have to do a whole lot. Usually a drive shows up, we slap it into the system, and it is good to go. 

They have been good working with our newer administrators, who are not as familiar with the storage platforms. They over take them, do the upgrades, or walk them through the deduplication processes. 

We can call them with anything. We also have a TAM who helps and facilitates a lot. Once we get to the back-end texts, we never have a hassle, even if they determine that it is a VMware or Cisco issue. It does not matter. They are always willing to stay on the phone, all we have to do is open a case with the other provider, and everybody works together and says,“Here is what we found.” 

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

Environments that I worked with have always been hodgepodge. We are not huge right, but HBMSUs were even older than when EVAs were popular, but those were limited to block storage only. Yet, in the organization I am in, they were limited to Fibre Channel, therefore going to a FlexPod and having the flexibility to do NFS, CIS to do Fibre Channel, ISCSI, etc., it doesn't matter. We can do it all in one array for whatever vendor solution that we pick, whatever storage they say we need, whatever hardware they say we need, we have it. We have the flexibility to put it all away.

How was the initial setup?

It was very straightforward. Each solution is a little bit different: Everything from the E-Series being the simplest to MetroCuster being the most complicated, but they have all been relatively straightforward to setup. We have been using NetApp services for most startups, so this has been a big benefit, especially with technology that we did not necessarily understand right off the bat. 

What about the implementation team?

Occasionally, we utilize support for upgrades. They do the prechecks, make sure the firmware is up-to-date, and run our baselines through to ensure everything is good. 

One of the NetApp consultants, Patrick Rodrigue, has always come out and helped us.

What was our ROI?

It is a little more difficult in government because we do not track that much on the soft dollar side. They look at it more as a capital investment. However, I can tell you from when I started there, when I started with the organization and we put our first FlexPod in our downtime has significantly been reduced.

From that prospective, if we look at our return on investment, we had have more productivity uptime. Our end users are obviously happier and IT is not constantly getting a black mark from the business because tech has not been worked.

What's my experience with pricing, setup cost, and licensing?

Make sure you understand the technology that you need, and anytime that you are buying any storage make sure you understand storage. Do not just buy storage based on what somebody sells you in terms of IO or throughput. Buy storage based on the solutions you need, the technologies you need, and what will make your life a lot easier down the road.

Which other solutions did I evaluate?

It is not so much that we need to invest in FlexPod. I work pretty closely with all of our vendors, and a lot of times, we look and we evaluate. We evaluate all the available solutions out there. It does not matter whether it is a FlexPod or if it is one of the illegal EMC counterparts. We evaluate them all. We look at everything from Nimble Suite and the big brands, like FlexPod. Every time we go out and we evaluate solutions with their flexibility. The flexibility of a FlexPod wins out every time.

Having an extremely cost effective solution which is a pain in the butt to manage, a pain in the butt to support, or overly complex does not really do us any good. It ends up just costing us, even though we do not track money. It ends up costing us time, which in turn, costs us money, and management does look at that. 

We look at performance. We look at the available options and how they unified a platform, especially when it comes to storage. Recently, we were comparing FAS units to a VNX from EMC. The big difference and big selling point for a FAS unit was the data filer with virtualized block put right on top of it. We do not have to maintain separate controllers. The VNX had to have a Solera and a clearing head in it in order to do block and file based storage. We had to separate discs at a point in time. This is a few years ago, so some of it has changed since then. However, when I talk about simplicity to manage, it also goes into cost. 

On the EMC side, I would have had to have dedicated disc per file and dedicated discs for block-based storage. On the FAS side, I could do whatever I wanted. I just had a big disk pool and I could divide it up however I wanted.

What other advice do I have?

We purchased through CDW. They were knowledgeable about the solution. They won the bid. It was very simple with us. We sent it out for a bid and they came back with the lowest cost on the response.

Most important criteria when selecting a vendor: Cost is always important, but it is not our base. We look at performance, availability, overall usability, and simplicity.

Disclosure: IT Central Station contacted the reviewer to collect the review and to validate authenticity. The reviewer was referred by the vendor, but the review is not subject to editing or approval by the vendor.

Add a Comment
1 Comment

author avatarAdriano-Simao
Top 5LeaderboardReal User

FLEXPOD is more reliable solution i could find. We are running since 2012 with Cisco UCS B Servers and FAS3000 Series on both sites without MetroCluster but never went down. Since that time, not more than 10 hard drives changed (~192 TB Raw). Now we are moving on to AFF Models.