Skip to main content

Title is most of the question… I have a 3-node, local-only cluster (no remote / stretch replication or anything). 

What do I gain from setting up Prism Central? It seems like Prism Element gives me all of the management capabilities I need (running AHV, probably won’t have more than 20-40 VMs).

It feels like standing up PC, even in the single-VM / non-redundant version, will just be a waste of RAM and storage on the cluster. Is there some compelling argument for Central that I’m overlooking because I’m a Nutanix noob?

Hi BradenMcG-DC,

I think it comes down more to what you want to do and what features to use. Starting with AOS 5.5, role-based access control, application management (Calm), micro-segmentation policies, and Prism Self Service (PSS) features management is handled by Prism Central. There are also Karbon and Objects.

If none of these interest you then it is possible that you do not need PC. There is nothing that says you must have it. It is a convenient way to manage medium to large environments and leverage non-basic features. But you know what you need.

You can always give it a try and then uninstall it later. If at some point you do grow and notice that managing your cluster is taking more time than you would like to commit you can deploy a PC.

I agree with you, having something that you don’t use is a waste of resources.


Hi BradenMcG-DC,

I think it comes down more to what you want to do and what features to use. Starting with AOS 5.5, role-based access control, application management (Calm), micro-segmentation policies, and Prism Self Service (PSS) features management is handled by Prism Central. There are also Karbon and Objects.

If none of these interest you then it is possible that you do not need PC. There is nothing that says you must have it. It is a convenient way to manage medium to large environments and leverage non-basic features. But you know what you need.

You can always give it a try and then uninstall it later. If at some point you do grow and notice that managing your cluster is taking more time than you would like to commit you can deploy a PC.

I agree with you, having something that you don’t use is a waste of resources.

Alona,

I appreciate the reply. For some background, I have “seen” and briefly interacted with Prism / Nutanix in the past at a prior job, but it was not my primary focus. So, I have a small amount of “experience” but no formal training or much in the way of “work with” Nutanix. 

In my current job, I pushed for Nutanix (over VMware + vSAN or HyperV) due to my former coworker’s excellent experience with your product.

At my new position, Nutanix is going to run all “back end services” for a small ISP. So, it’s primarily going to host Linux or BSD VMs, and there won’t be much in the way of “self service” or much with privilege separation/RBAC (the few people who will have access should be ok with full access). We have 3x1175S-G7 nodes (hybrid storage) with 128GB RAM each, and I don’t expect more than 30-50 VMs in total (and even that may be a stretch). So, I’m not keen to “waste” another 26-30GB of RAM for Prism Central if it isn’t going to do much for us. You raise a good point though, I could always stand it up and and play around to see if it gains anything, and then take it back down if it isn’t helpful.

Part of why I suggested Nutanix was that I know how easily it can be scaled. :wink:  We are starting small, but should we have the sort of growth that our C-levels want, I can envision a scenario where we do need to scale out, and that’s much easier to do with this product than others on the market (IMO).


Thank you for sharing the background and thank you for such positive feedback!