FlexPod Review
It is easy to use. If you follow the reference document how to set it up, it provides a stable environment.

Valuable Features

One of the valuable features is ease of use. Getting any environment set up is probably the easiest thing to do. You can set up the entire solution in about a day or so. When we have a requirement for a specific project, we don't need to worry about getting into different gears. FlexPod is a converged infrastructure, so when you get it, you have reference architecture. You just install it and start using it. Those kinds of features are really good.

Improvements to My Organization

The storage scales out and you can keep on adding your UCSs. Adding the whole scale-out technology is great. You can grow as you need to and that's a really good feature.

Room for Improvement

I don't think there's much to be improved with the tool since you can now scale out storage. Before that, this was a shortcoming in that you had to upgrade the head every time.

I would like to see the ability to combine a couple of FlexPods into a cluster. You cannot do that now. You cannot combine two FlexPods into a single entity, into a larger FlexPod. To the best of my knowledge, FlexPods are meant to be in silos and you cannot create clusters at all. If there is a way to do that, that would be interesting.

If there could be a FlexPod management piece, then you could manage all your FlexPods from a single console. That piece is missing even though there are some NetApp tools where you can manage. However, those management tools are specific for the storage.

I would like to be able to manage FlexPod as a single entity for all the different components. If there could be a single tool which can monitor all of them together, that would definitely give a big edge. It would be great if you could manage all of your FlexPods from a single location.

Stability Issues

The tool is pretty stable. Stability-wise, I would give it the highest rating. If you follow the reference document, in terms of how to set up FlexPod, it's a very stable environment upon which to work.

Customer Service and Technical Support

I have used technical support, but not exclusively for FlexPod; maybe questions here and there related to the FlexPod environment. I don't think we have ever used FlexPod tech support which is there in NetApp. We have pretty competent resources in-house, so we never feel the need to use FlexPod support.

Previous Solutions

I was involved in the decision to switch to this product. We were looking for a tool that was designed for the way our organization works. We wanted a silo environment for different applications. Since we have segmentation in our company, we have different domains, and FlexPod really does fit in really well in those situations where you need a FlexPod for a particular application or for a job area. There’s an idea of implementing Citrix and VDI on it, so those kinds of applications are really good.

We were the first company to use EMC's Vblock implementation, and it was a Vblock pain. I was not there when the company selected Vblock, but I was told that there were a lot of issues. Being the first customer on Vblock was really a nightmare. We had to move to FlexPod. But it doesn't mean that Vblock was not good. Our timing on the purchase of Vblock was not right. Our expertise in the company was more Cisco driven and FlexPod really fit in well with that.

Initial Setup

I was involved in the installation. FlexPod, or any converged or hyper-converged infrastructure, requires a lot of planning. Once you have your planning done properly, you can just work with networking and other teams. If you have a good coordination with the teams, it's pretty easy to set-up.

Other Solutions Considered

I was not involved in the decision-making process. Things have changed since Vblock was launched seven years ago. FlexPod and Vblock both have very similar architecture and I don't see any big pros and cons between them. I think it's just a comfort level with respective companies. If a company has more investment in Cisco and VMware, that's how the FlexPod architecture is designed. I have no comment on Vblock right now.

There were no other vendors at the time. I was going with NetApp only for non-FlexPod environments. That was when we started buying stuff, which was about six years back when there was no competition. However, everybody has their own FlexPod now. Nimble has something like their own stack. Pure has a Pure stack. Everybody's coming with their own converged infrastructure and we are looking around.

When selecting a vendor, partnership plays an important role. A good partner will provide a kind of an independent review of the different vendors. When we select a vendor, we look at:

  • Our means
  • Our relationship with the vendor
  • The standing of the vendor in the industry
  • The vendor's new innovative technology
  • How the vendor is competing in the market
  • How competitive the vendor is in terms of price.

We look at other technologies because other technologies do provide similar kinds of things as NetApp at a cheaper price. That's how other vendors are rolling over each other in the market right now. They can provide the same thing for less money. These are important things, but the company stability and their goodwill in the industry are important factors as well.

Other Advice

Our experience using this tool is that we have been very happy with it for over six year. The solution has given us whatever our company has wanted. It has delivered in a very short time and has quick turn-around for different projects.

I also suggest looking around. NetApp is a good case for us. It really solves our issues. Although there are other solutions available on the market, this tool is definitely worth looking at it.

FlexPod is not cheap and the way things are going, you could probably get the same thing at half the price from another vendor. NetApp has to be very competitive on the prices in order to really compete in this market.

Disclosure: I am a real user, and this review is based on my own experience and opinions.

Request Demo

Learn more about Flexpod

1 Comment

Greetings from a VBlock owner and also a NetApp shop that had the first of the 300 series ever delivered. I Had serial number 1 and 2. Each one was parked in a data center within the metro but are separated enough to not likely be in a common major disaster unless the whole city is involved and if that happens there are bigger problems. Anyway back to some of your comments and my background.

I Have a storage specialist that watches the VNX and VPlex as well as NetApp and other storage systems. I oversee the compute and networking within the two current and now retired older VBlocks and have a good knowledge and comfort with the overall network systems, core switches and understanding of the metro 10 GIG LAN between our offices and the two data centers.

A few years ago we installed the first set of two VBlocks that were separate islands but we used the EMC RecoverPoint in place to replicate the data between the two data centers in near-real-time copies at both ends. This does require doubling of storage but that was our initial DR strategy. If one site was lost we brought up the system on the other side. Luckily this never was needed.

Later we added additional equipment to make the two VBlock's into a more high availability setup with VPlex to keep both VNX's in Sync. Since our two data centers are within the metro area and we had redundant 10 GIG between them we could do synchronous rather than async writes to both sides. On the LAN we did OTV with stretched layer 2 / 3. We set up VCHeartbeat with redundant VCenters for HA on the VCenter between the AMPs. The whole environment was switched over from one site to the other at least once during their lifetime as we did an in-place upgrade of the VNX's and by VMotioning between the two VBlocks we had little to no end user outage. Running VMware 5.x but could not upgrade to VM 6 due to hardware incompatibility issues and age.

When the OLD VB-300's hit EOL we migrated the VM's on them to two new VB-340's, one landing in a NEW data center that we were moving to. We migrated data and VM's between the old and new VBlocks using VPlex connections between the old and new VNX systems to sync the storage and some VM scripting with some assistance from a VCE consultant that moved in bulk migrations of VM's. Most of which only took a short shutdown on the old system and pull in and power up on the new VBlock. Not much more than a scripted reboot that also performed some cleaned up of old VM hardware, fixed tools and removed old floppy disks.

The two new VB-340's have their own separate VCenter 6 manager servers but are in a common VMware domain so they can both see each other in the browser client and can on the fly VMotion between the two VBlocks since they both see each other's disk drives via VPlex and OTV, All works well.

Now for not able to "Cluster" two systems is more a matter of implementation and how close the two VBlocks / FlexPods are for the right tools for replication between the two storage systems. If you are doing snapshots from one NetApp or other Storage System under the FlexPod solution it is a matter of how frequently they are synced up. I don't thing NetApp has the ability to directly do a metro synchronous write between two NetApp HA system but it may even be possible to implement Cisco VPlex to present the disk LUNS to the VM hosts and keep the storage in sync if they are close enough to do synchronous writes to the storage via VPlex. OTV solves the networking. Its just a matter of applying the right tool for the job.

Like (0)25 May 17
Why do you like it?

Sign Up with Email