Why Raw Calculated Storage Is Not the Same as Purchasable Drive Capacity
A storage calculation based purely on bitrate, retention period, and camera count produces a raw storage requirement — but the actual physical drive array a facility purchases and installs needs meaningfully more raw drive capacity than this calculated figure, because RAID (Redundant Array of Independent Disks) configurations, used to protect against drive failure without data loss, dedicate some portion of total physical drive capacity to parity or redundancy information rather than usable video storage.
How RAID 5 Parity Overhead Works
RAID 5 distributes parity information (data that allows reconstruction of any single failed drive's content from the remaining drives) across all drives in the array, with the effective overhead equal to one drive's worth of capacity regardless of how many total drives are in the array. For a 5-drive RAID 5 array using identical drives, four drives' worth of capacity is usable for actual data, and one drive's worth is effectively consumed by parity — meaning usable capacity is 80 percent of total raw capacity for that specific array size, though the overhead percentage varies with total drive count (a smaller array has a larger overhead percentage, since one parity drive is a larger fraction of a smaller total).
How RAID 6 Provides More Protection at More Overhead Cost
RAID 6 extends the same underlying concept with two independent parity calculations distributed across the array, allowing the array to survive TWO simultaneous drive failures without data loss, compared to RAID 5's single-failure tolerance. This additional protection comes at a correspondingly higher overhead cost — two drives' worth of capacity dedicated to parity rather than one, commonly resulting in overhead in the range of 33 to 40 percent for typical surveillance array sizes, compared to RAID 5's roughly 20 to 25 percent for similar array sizes.
Why the Choice Between RAID 5 and RAID 6 Matters More for Larger Surveillance Arrays
As surveillance storage arrays grow larger — more drives, larger individual drive capacities — the statistical probability of a second drive failure occurring during the rebuild process after a first drive failure increases meaningfully, since larger arrays with larger individual drives take longer to rebuild, extending the vulnerability window during which a second failure would cause genuine data loss under RAID 5's single-failure tolerance. This is why RAID 6 is often specifically recommended for larger surveillance arrays despite its higher overhead cost — the additional protection becomes proportionally more valuable as array size and rebuild time both increase.
Why Filesystem and Additional Headroom Overhead Also Factor In
Beyond RAID parity overhead specifically, real storage systems also lose some capacity to filesystem overhead (metadata, formatting overhead inherent to how a filesystem organizes and tracks stored data) and typically benefit from additional headroom margin beyond the bare calculated minimum, to accommodate unplanned retention extensions, temporary bitrate spikes from unusually active scenes, or gradual system growth before the next planned capacity upgrade. This is why a complete storage sizing margin commonly combines RAID parity overhead with some additional headroom percentage, rather than accounting for RAID parity alone as if it were the only source of overhead.
Why This Site's Calculator Includes a Combined Margin Input
This site's Video Storage Sizing Tool applies its margin input as a combined factor covering both RAID parity overhead and general headroom, reflecting how real storage specifications are actually built in practice — a 20 percent margin, for example, might represent RAID 5 parity overhead on a moderate-sized array plus a modest additional headroom buffer, while a larger array or a RAID 6 configuration would reasonably call for a higher margin percentage to accurately reflect its greater actual overhead.
Why a Hot Spare Drive Adds Further Consideration Beyond Parity
Many production surveillance storage deployments also include a hot spare drive — an additional drive kept ready and unused during normal operation, which the RAID controller can automatically activate to begin rebuilding the array immediately upon a drive failure, without waiting for manual drive replacement. A hot spare adds further physical drive count beyond the data-plus-parity drive count, though it does not itself consume ongoing usable storage capacity the way parity drives do, since it remains idle and reserved until actually needed.