One of the most valuable features is that I have one vendor that I have to contact to go to all my vendors. I don’t have to call Cisco, NetApp, VMware. I just make one call to one of them, they bridge all of that for me; that's nice.
Also, with the documented architecture, it's not something that we're testing right off the bat; it's been proven and it works.
Improvements to My Organization:
It allowed us to streamline support tickets that come in. We don't have to call three different vendors. I just call one and they take care of all of that for us. It's been very helpful.
Each individual solution has its own cost benefit. It really fit within our organization. We had a lot of the existing technology there. We were just missing a couple pieces of it. So, once we got those pieces, we were able to certify it with FlexPod. We didn't need to buy too many new pieces. It fit into our original architecture.
We have cut down on the management team running FlexPod compared to our previous system. We've reallocated one FTE so far because of it. We can do more things with fewer people on the team.
Room for Improvement:
I know there's some new Cisco stuff coming down the road that we might be looking at. UCS Minis: I know that they're going to be supported right off the bat.
Nothing's perfect. There's always room to increase: more hyper-conversions; smaller form factor is always on our mind; better ways to align disk up for us; how can we split off our disk correctly for each HA pair that we have, from a cluster standpoint.
There was a little mixup with, at least NetApp, coming to the market with flash. They've slowly gained ground in that marketplace; I’m waiting to see how that plays out. I know Mars was trying to be a big hit for them and then when they dissolved, that kind of set me back a little bit from a time standpoint. Mars was their all-flash platform; something separate from WAFL that we're doing with ONTAP. That had me looking at that infrastructure. That's the only reason why they lost a few points in my rating.
Use of Solution:
I have been using it for about three-and-a-half years.
It is a stable infrastructure. We've haven't had any major outages in the past three-and-a-half years since implementing it. There's been no downtime from a hardware standpoint that we weren't able to address quickly.
It is very easy to scale out the infrastructure, add more pieces to it as we needed; just kind of plug and go. That’s very easy.
That depends on when I call. Cisco is a little bit more difficult than some of the other ones, but calling it up has been great. They've bridged that gap a couple times at VMware as well.
We were using a mixture of Dell and HP solutions. We were using Dell 910s for a lot of our ESX environment. We were using old Cisco MDS switchers for fiber channel. We were able to consolidate all that infrastructure down and use one standard platform coming off the Nexus.
My engineers worked with one of our strategic partners to help implement the solution. It has been, from a design standpoint, much easier to get set up and running, and much faster than doing it outside of FlexPod.
We haven’t had any technical issues with getting it set up or with running it; none at all. I'm very happy.
Cost and Licensing Advice:
It's a little more expensive to go down the FlexPod route, but I think the ease of management in having all the vendors aligned really helps us in the long run. There is more upfront cost, but less down the road that we have to pay; maintenance and support, man-hours, actually managing the system.
Other Solutions Considered:
We already had NetApp in-house. It was very easy for us to use at least that as a storage platform, so it was just finalizing on the Cisco UCS part. We needed to come up with a hardware platform that we could use and UCS was the hardware platform.
HP’s BL series blade was the other one we were looking at from a blade standpoint.
We decided to go with FlexPod instead of HP because we were already a leveraged Cisco partner, a Cisco shop, with all of our route switch and all of our data center core switching. It was very easy for us, then, to assimilate the UCS chassis within our existing infrastructure without any other type of complexity.
Make sure you get all your requirements up front. Make sure that the protocols that you want to use are supported by your vendor. There are a lot of niche players out there that will say they will do something. When you get them onsite, they don't or they don't perform as well. We were looking at a couple of other flash providers, including Nutanix and Tintri.
Bringing them onsite for a demo works great, but when you start talking to people who actually use the solution... We found that in the demos, they couldn't live up to the promises they were making, or it didn't make sense to add more infrastructure in. I can't get rid of my core NetApp infrastructure, so it didn’t really make any sense to add another storage vendor in, and increase the complexity. Using FlexPod has been simpler than adding in another vendor.
NetApp might be a little more expensive but in the long run, it pays off for itself.
I’ve been very happy with their technical ability, their technical delivery, and the usability; it’s very easy.
Disclosure: I am a real user, and this review is based on my own experience and opinions.
Nov 22 2016