FlexPod Review

It can scale, compute, and storage independently by what we need


What is our primary use case?

We use FlexPod for everything: Running our virtual stack, all our research data, etc.

How has it helped my organization?

Moving from rack and stack servers (Dell EMC and HPE) to having an overall encompassing design with UCS, NetApp, and VMware, made us more resilient. We can lose nodes and drives and also stuff can go down, but there is no downtime. We can recover quickly. 

It makes disaster recovery (DR) easier as well, if you have a FlexPod set up in one place, then add a DR set.

What is most valuable?

It can scale, compute, and storage independently by what we need. As opposed to in the hyper-converged realm, you are sort of locked into a linear growth pattern.

What needs improvement?

I would like them to simplify the UCS configuration. I appreciate that they have about a billion options and a million switches that you can mess with, but this creates a lot of confusion sometimes. I feel like you almost need a Master's course to figure out what you're doing with UCS.

What do I think about the stability of the solution?

It's extremely stable.

The solution is resilient. We have suffered failures before without any downtime.

What do I think about the scalability of the solution?

We are constantly scaling. I just added half a petabyte of storage not too long ago to the storage site. Adding new nodes and making new UCS clusters allows us to scale any way that we want.

How is customer service and technical support?

  • With NetApp, technical support has always been great. 
  • With Cisco, it depends. 
  • VMware is horrible. I hate calling them for anything.

Which solutions did we use previously?

We were running on legacy rack and stack; just single servers doing single things with server sprawl and multiple racks of servers. It's not a great way to do things. That's what drove us to FlexPod.

How was the initial setup?

The initial setup was complex. UCS is not the easiest thing to configure from the ground up. The networking pieces can get confusing, especially when you are talking about virtual segmentation. It is not as easy as other things now on the market, such as hyperconverged.

What was our ROI?

Coming from a rack and stack server model to FlexPod, it has saved us a lot of time (approximately hundreds to thousands of hours).

Which other solutions did I evaluate?

We also looked at VxBlock from Dell EMC.

What other advice do I have?

If you need to scale, storage and commute independently, then you want to go FlexPod. If you don't have that sort of need and want something simple and easy to throw up and use, despite some of its shortcomings, hyper-converged is probably the way to go. It really depends on how big you are and what you need.

Versatility is great. However, in this day and age, it is probably more complex than it needs to be, especially on the Cisco side. I am not a huge Cisco lover. UCS is getting long in the tooth. It's great for what it is, but it is now overly complex compared to other solutions on the market.

FlexPod was at one point on the bleeding edge. Now, I think the bleeding edge is hyperconverged, and I know Cisco and NetApp are looking into that independently.

We use FlexPod for Managed Private Cloud, which is great.

I don't love the Cloud. It is a good space for second copy backups and maybe bursting into the cloud depending on what your application workload is like. However, I'm not a lover of the hybrid cloud model, or even going fully into the cloud, unless you are willing to undertake the paradigm of creating your applications and workload for it. Moving your legacy info into the cloud is expensive and a bad move.

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

Stay connected with Flexpod

Add a Comment
Guest

Sign Up with Email