So-called "NAS" drives, is mostly marketing wank aimed at the consumer market. You really need to consider the application (hardware RAID vs software RAID), size of the system and the specifications of the drive.
NAS drives typically (but not always) have a error recovery control enabled (also known as TLER or CCTL depending on the manufacturer), or it's switchable through software. This basically limits the time taken for error recovery. With this enabled, if the drive encounters a read/write error, it will give up after a shorter period and report the failure back to the HBA/RAID controller to deal with. Typically, this timeout is shorter than the timeout of the RAID controller. In drives with no ERC limit set, it may try continually (and for long periods) to recover from the error, which presents as a unresponsive/failed disk to the RAID controller, which in effect will cause it to drop out of the array.
Manufacturers often claim that NAS drives are more reliable and are designed for 24/7 operation, however in all my years in the industry, I haven't come across a disk yet that isn't "designed" for all day, everyday operation. You may want to consult the MTBF or TBW specification for the drive(s) you're looking at, to give you some idea.
Some drives (even ones labeled suitable for NAS) use shingled magnetic recording (SMR). This was/is particularly true for some of the WD Red/Blue/Black and Seagate Barracuda and Skyhawk drives. Whilst these might not present much of an issue or a performance hit in smaller NAS systems, but as arrays grow, they can be a real bottleneck, particularly with some of the drive-managed SMR (DM-SMR) drives like the WD Red's. (As far as I'm aware, the Seagate IronWolf series uses CMR/PMR exclusively).
I stopped using consumer-grade drives about 15-20 years ago, even for my home NAS setup. They were just too much of a headache. However, these days, traditional hardware RAID is basically considered "old-school" in favour of better, and more resilient software-based setups (like ZFS), so consumer drives may be less of an issue (and more tolerated), but do your research and plan your system properly.
Even though significantly more expensive, I only buy enterprise grade drives (like the WD/Hitachi Ultrastar's). Even second-hand, they've proven to be a very reliable drive for many years. I had an array of 2TB drives running continuously for over 10 years and are still good today (although I retired them for larger drives).
As I always say when it comes to IT/Technology:
- There is no one-size-fits-all solution.
- You generally get what you pay for.
- If something appears "cheap" there is usually reason for it, which might not be favourable for you.
- There is more than one right way to do things.
- Consider what's most important to you: Cost vs. Performance, vs. Reliability; Pick two.