In this specific model, it could be beneficial to have a server configuration similar to Cisco, like the kind offered in HPE Superdome Flex X280. The ability to separate the chassis is important because we are forced to use two chassis even with lower loads.

So introducing more flexibility in the Superdome environment would be helpful. 

One other problem is energy consumption. We are forced to use both chassis even when we only need, for example, all four processors. It's not always used at full capacity every day. 

When underutilized, we still need to have both chassis powered, which creates an energy consumption issue.

While the machine can be configured to reduce consumption, the idea of having different possible configurations with varying chassis usage, similar to the Flex version, could be interesting for this model, too.

This model leverages specific features of the Intel MAP board, so switching like in the Flex might be difficult due to the motherboard switching involved.

HPE Superdome X can improve by adding a lot of cloud capabilities to allow this solution to be cloud-ready in case the customer wants to move it to the cloud.

This is just my wild speculation, I'd like to see the onboarding of some storage class memory to really expand the already very large RAM, into something that could be even much bigger. This could really help in solving the problem of when you want to do work on a node that's 12 terabytes of RAM and getting the data on/off can be a limiting factor. Thus, having a richer storage hierarchy within the node could actually have some interesting capabilities.

The price of this product should be cheaper.

It would be helpful if you could do the maintenance completely online.

Looking back to our original purchase of Superdome X, we bought the smallest model. Now, the newer models have more flexibility and power, and we're a little upset at ourselves that we didn't purchase the next level model. So, based on the model that we do have, I'd like more flexibility as we're only able to extend one blade for one to five kilobytes. Needless to say, this is getting too small for us.

It always comes down to OS. I know we support Windows and I think it's Red Hat, and I'd like to see more OSs supported.

I know HPE has another product, called Synergy, which is similar to Superdome, but I would think, if they can add maybe storage, even if it's a mini-set of storage to each, which can be used externally, they should do it.

There's a big upfront investment, but the total cost of ownership should be cheaper over the long term.

What would make it better from my point of view is if HPE spent more time on testing with the actual built-in Red Hat Linux drivers, as opposed to always trying to say, "Use our driver."

Hardware is hardware, there might be an issue with memory they have to replace memory or something but nothing major. There is some reporting that we will probably want to use it for on our next projects. We also use this for our major databases but we don't virtualize it.