RAID Storage Capacity Estimate Calculator

Three things reduce it. First, redundancy: every RAID level except RAID 0 reserves some capacity to protect your data, which is what lets the array survive a drive failure. Second, how capacity is measured: drive capacity is quoted in decimal units where 1 TB is one trillion bytes, while operating systems usually report in binary units, so a 4 TB drive shows as roughly 3.64 TB. Third, formatting: the file system itself uses a small amount of space. The calculator shows usable capacity after redundancy; expect your operating system to report a slightly lower number again.

It depends on how you want to balance protection, capacity and speed. RAID 1 is the simplest protection for a two-drive setup. RAID 5 offers a good balance from three drives up, surviving one drive failure while keeping most of your capacity. RAID 6 suits larger arrays because it survives two simultaneous failures. RAID 10 favors performance alongside redundancy, at the cost of half your raw capacity. RAID 0 offers no protection at all and should only be used for data you can afford to lose. Whichever you choose, use drives rated for continuous RAID operation.

Yes, for standard RAID levels. If your NAS is configured as RAID 1, 5, 6 or 10, the capacity math is the same whichever brand of enclosure you use. Some systems add their own formats: Synology Hybrid RAID (SHR) and ZFS RAIDZ, used by TrueNAS, allocate space differently and are especially good at reclaiming capacity when you mix drives of different sizes. For those, treat this result as a baseline and check your system's own tool for an exact figure. Whatever the enclosure, populate it with drives built for 24/7 operation such as WD Red Plus or WD Red Pro.

You can, but standard RAID levels size the array to the smallest drive, so the extra capacity on larger drives goes unused. Two 4 TB drives and two 12 TB drives in RAID 5 will be treated as four 4 TB drives, wasting 16 TB. For best results use drives of the same capacity and model across the array. The exception is JBOD or JBOF, where each drive is used independently and no capacity is lost, though you also get no redundancy.