New
#11
All my external HDDs are regular 3.5 inch models. Some of my external HDDs are the "green" type, i.e. the exact opposite of NAS/enterprise models. Not all green drives belong in the "slowest of the slow" ballpark, so I saw no real reason for me to avoid buying them altogether, and, instead, I went for a more nuanced kind of strategy, i.e., not all of my data was going to be accessed frequently enough to be able to justify the added cost if the price difference was fairly big compared to the difference in performance. All the other ones I have are just normal consumer models (7200rpm) the primary focus of which was always storage capacity vs cost while still not sacrificing too much on performance vs cost. I don't own a NAS, as empty NAS boxes are expensive, and they don't have any storaga capacity whatsoever because... well, they're empty! As for RAID 0 arrays and any other type of configuration that requires multiple drives to be kept spinning the entire time, I have no use for them. Keeping multiple drives spinning consumes more power, running them 24/7 limits choices because not all drives are designed to be reliable enough for that so it tends to hold back storage capacity vs cost, and I would need to buy one or more separate enclosures so the total cost would be much too high. (Even, if I decided to build my own storage pod.) I own a 16-port USB 3.0 hub that uses its own external 100 watts power supply. External HDDs also give the benefit of being designed to be portable so they are better protected when they get moved around a lot. A safe backup isn't safe until it is stored in a safe location away from the computer and physically disconnected from anything that is still connected to the power grid, and their integrity has to be verified often if you want to call them reliable so moving them on a regular basis tends to be somewhat of a compulsory type thing. What a lot of people seem to be hopelessly unaware of is that increasing the number of backup copies is what helps to avoid suffering permanent data loss orders of magnitude more effectively than choosing a drive that is touted to be the bees knees of reliability. (Before anyone should ask, MTBF stands for Mean Time Between Failure, not Mean Time Before Failure─as I saw a discussion about that in another thread not too long ago... )