← Physical Security Studio
📹

Video Storage Sizing Tool

CCTV / VMS · Recording Bandwidth · Retention

When to use: Use this tool when sizing the NVR/VMS storage array and network for a surveillance system. It estimates per-camera bitrate, total storage for a target retention period, the aggregate recording bandwidth the NIC and switch must carry, and compares H.264 vs H.265 vs H.265+ so you can quantify storage savings. Bitrate estimates are approximate — scene complexity and motion dominate real-world usage, so always verify against your camera or VMS bitrate calculator.

System Parameters
cams
fps
days
Parity + spare
%
Recommended Drive Configuration

At 8 TB surveillance-rated drives, the 6.22 TB target needs about 1 data drive, or 2 drives with one parity drive (RAID 5). Add a hot spare for critical systems and round up to the next array size.

Total Storage (with 20% margin)
6.22
TB required · 5.18 TB raw
Results
Per-Camera Bitrate1.00 Mbps
Duty Cycle (recording)100%
Aggregate Bandwidth16.0 Mbps (0.02 Gbps)
Raw Storage5.18 TB
With Margin6.22 TB
Retention30 days
Size the recorder NIC and switch uplinks above the aggregate bandwidth — a 1 Gbps link saturates near 700 Mbps of sustained writes. Use 10 GbE for large arrays.
Codec Comparison (same inputs)
Codec
Total TB
Savings vs H.264
H.264
10.37 TB
H.265 / HEVC ·
5.18 TB
50%
H.265+ (smart)
3.63 TB
65%
Storage is calculated for the same camera count, resolution, frame rate, duty cycle, retention, and margin under each codec, so the only variable is compression efficiency.
References & Notes
Bitrate estimates are approximate — scene complexity and motion dominate real-world usage (CBR/VBR).
H.265 / HEVC typically saves about 50% storage vs H.264 at equal quality.
H.265+ / smart codec typically saves about 60–70% vs H.264 on low-motion scenes.
Storage uses decimal units: GB = Mbps × 86400 × duty × days ÷ 8 ÷ 1000.
Retention drives raw capacity; RAID parity and headroom add usable margin.
Always verify against the camera or VMS bitrate calculator before final sizing.