At the end of the day it's still just a server. And servers are being tested.
This is true - but tested using standard and common methods (to include standard and common software) so the results can be compared with other "similar" servers using those same standard and common methods. You asked about this NAS running a specific program - a program that would NOT be considered standard or commonly used for testing or comparative analysis.
And servers generally are NOT used to crunch or process data. They are meant to "serve" data to attached workstations that then do the crunching and modifying of those data files that are then saved back on the server. Servers are meant to facilitate sharing of those files. NAS, after all, stands for "network attached
storage".
But with these basic NAS solutions the CPU is a limiting factor more than it is in models costing twice as much.
Yes, the CPU is "a" limiting limiting fact - but "a", in this case means it is just "1" of many limiting factors. It really takes very little CPU horsepower to fetch and save data to a drive. For sure, in many server scenarios, having more server memory is more important than having a faster CPU.
They aren't testing performance - just guessing it.
They are not guessing, they measured. But it was not meant to be a benchmark or a comparative analysis. As I said above, I am not surprised you can't fine what you are looking for. You are asking for something too specific that probably does not exist.
You said so yourself in your opening post, you could not find any benchmark using that database program. Don't be surprised if no one else can either.
If I could help you, I would.
The best I can do is suggest you narrow your search with
Google. You probably should set a budget too, and decide now if looking to use hard drives or SSDs, how many drives you want, if RAID or not, and the total storage capacity you need.