548 words
3 minutes
A massively overcomplicated NAS build

A while back I realized that I had actually completely run out of storage; my combined 5TB across multiple drives had completely filled up. Having become a data hoarder myself, I knew that I had to go out and buy more storage at once. After all, the biggest sin you can commit is deleting data you aren’t absolutely sure you will never need again. If you do, eventually you will realize that you do actually need it again, whether it be immediately afterward, the next day, or even a few months down the road.

But just buying more storage is much easier said than done. My main workstation machine had actually completely used up all the available SATA ports (at least, the ones that actually worked), and I knew that I was going to have to either buy a larger drive and move all my data over, or build an entirely new NAS. One of my biggest priorities going into this project was my interest in setting up a home streaming server, that being a combination of [Jellyfin], [Stremio], managing torrents and other miscellaneous services. Given that our ISP had just recently upgraded us for free to a better plan, this meant that I could finally host a Minecraft server without being our up-link being saturated, and causing network lag. Having so many requirements meant that my priorities when picking out the components for my NAS/server combo was quite different from a traditional NAS build.

Please note that I don’t have any prior experience to build a NAS, and as such don’t really know what I am doing. Each component that I picked out was very specialized to my exact needs, and this shouldn’t be taken as a recommendation for you to exactly do what I did!

Cases#

Another one of my main priorities while designing this build was ensuring that this NAS could be upgradable in the future. This meant that any pre-built NAS, such as from Synology, was immediately out of the question. I wanted enough computing power to run a myriad of services, and enough drive bays to support 100TB+ if it ever came down to it.

Eventually, I landed on 3D printing my own case. Pre-made cases, such as the [Fractal Design’s Define] series was just too expensive when a much cheaper alternative was available. We had previously bought a Prusa MK3S+ printer a couple of years prior, and I hadn’t really used it enough to justify the purchase. This way, not only would I minimize costs, but it would also give me a reason to finally really get into 3D printing.

After searching around for a bit, I eventually decided on printing this [12-Bay ATX NAS case] design by [makerunit]. The reason behind choosing an ATX build is that I would be able to reuse my current hardware from my workstation, that was also an ATX build. This would also give me enough flexibility to potentially add in a GPU to accelerate transcoding (AMD’s CPUs don’t support Intel’s QuickSync), a 10GB NIC, HBA cards, and still give me space for things far into the future.

This part of the project ironically took the longest, mostly because of actually having to learn how to print and maintain my printer…

[//] # (https://www.youtube.com/watch?v=h7FyS37KHms)

[Jellyfin] https://jellyfin.org/

[Stremio] https://www.stremio.com/downloads

[Fractal Design’s Define] https://www.fractal-design.com/products/cases/define

[12-Bay ATX NAS case] https://www.printables.com/model/1278847-12-bay-atx-nas-case/comments

[makerunit] https://www.youtube.com/@makerunit

A massively overcomplicated NAS build
https://blog.rushii.dev/posts/a-12-bay-overkill-nas/
Author
rushii
Published at
2025-07-07
License
CC BY-NC-SA 4.0