Frigate Recording Storage Guide: Size HDD Capacity

Frigate recording capacity planning

Frigate storage planning becomes straightforward once the camera record stream is treated as a measurable data source. Instead of guessing a number of terabytes from camera resolution alone, calculate from the average record bitrate, how often footage is retained, and how long it must remain available. This guide walks through that math, shows current recording-storage products near the top, and separates raw footage size from the extra capacity needed for filesystem overhead, growth, redundancy and operational headroom.

See live hardware Size my workload

Quick answer

What this Frigate workload needs

For continuous recording, storage grows almost linearly with aggregate camera bitrate and retention days. For event-focused recording, multiply by a realistic duty cycle rather than assuming all cameras are quiet. Frigate writes normal recordings directly from the camera stream without re-encoding, so camera bitrate is the most useful starting input for the storage planner.

Live Amazon hardware

Current products that fit this decision

Compare current Amazon listings relevant to this guide. Product availability and prices can change.

Checking the dedicated Frigate Storage catalogue…

Buying decision

Match the hardware to the bottleneck

Use the calculator to establish the footage requirement first. Then decide local HDD, SSD, or network storage, and finally decide whether you need redundancy. Reversing that order often leads to buying a RAID layout before anyone knows how many terabytes of video the system actually produces.

Interactive decision tool

Frigate Recording Storage Planner

Use the workload inputs to size a practical Frigate hardware tier before comparing live products. Results are planning guidance; verify current Frigate support, camera bitrates and the exact hardware configuration before deployment.

Compatibility checklist

Four checks before you purchase

Keep live products near the decision

Current Amazon listings appear above the long guide so capacity, interfaces and prices can be checked before deep reading.

Measure the real workload

Use camera bitrate, retention, PoE draw and network aggregation instead of relying on fixed per-camera rules.

Separate application and recording storage

Keep the Frigate configuration/database role distinct from bulk video retention so each storage tier can be sized correctly.

Verify the complete data path

Camera, switch, NVR, storage and UPS are one recording system; a weak link anywhere can affect reliability.

Use Mbps instead of megapixels for capacity math

Resolution affects bitrate, but it does not determine it by itself. A 4K camera set to 6 Mbps may consume less storage than a noisier 1440p stream running at 10 Mbps. Read the configured record bitrate or measure actual traffic, then total the cameras that share similar settings. This converts storage sizing from a marketing-spec exercise into a repeatable calculation that can be checked after deployment.

Understand what Frigate writes to the recording path

Frigate documentation says recordings are written directly from the camera stream without re-encoding in normal operation. Recording segments are cached and retained according to configured policies. That means the record stream codec and bitrate influence both capacity and playback compatibility. The detector stream may be lower resolution and lower bitrate, but it should not be used as the storage input unless that same stream is intentionally assigned to recording.

Continuous recording is the simple upper-bound case

For a conservative estimate, assume every camera records 100 percent of the day. Aggregate Mbps can be converted into bytes per second, then multiplied by 86,400 seconds and retention days. This is useful for sites where continuous evidence matters or where motion rules should not create gaps. It also gives an upper bound that makes it easy to see how much storage event-focused retention could save.

Event recording needs a realistic duty cycle

Event-focused storage can be much smaller, but a guessed 10 percent duty cycle may be dangerously optimistic for a driveway, shop floor or street. Estimate activity separately for different camera groups. A quiet indoor camera can have a low retained percentage while an entrance camera may be active for much of the day. When in doubt, start conservative and tighten retention after observing real storage growth.

Retention policy should reflect the evidence window

Longer retention is not always better if it forces the system onto slow or fragile storage. Decide how far back you genuinely need to review incidents. Homes may choose one to four weeks, while business or compliance needs can be different. The correct number is operational, not technical. Once retention is explicit, the capacity calculation can drive a rational drive purchase rather than an arbitrary largest-disk decision.

Add free-space headroom after the raw footage calculation

Raw video size is only the first number. Filesystem overhead, temporary operations, bitrate spikes and future camera changes all consume capacity. A practical planner adds 15 to 25 percent headroom after the footage estimate. If the system is expected to grow, add the future cameras explicitly rather than hiding all growth inside an enormous safety percentage.

Local storage usually produces the simplest write path

A local HDD or SSD keeps recording I/O on the Frigate host and avoids dependence on a network mount. For a small dedicated recorder, that simplicity has real value. If you use a NAS, make sure the network, share protocol and mount behavior can sustain the write workload. Frigate troubleshooting documentation notes that network-share performance and mount options can affect recording copy times.

Network storage should be tested, not assumed

A NAS with fast disks can still become the weak link if the connection drops, DNS fails, SMB/NFS behavior is poor, or the host routes recording traffic through a congested uplink. Measure copy times and monitor errors before trusting a remote target. If the NAS is several switches away, include those devices and their UPS coverage in the reliability design because the recording path is only as strong as its dependencies.

Storage speed and CPU pressure can interact

Slow recording copy times are not always caused by the disk itself. Frigate troubleshooting guidance also suggests checking CPU pressure when the storage path otherwise appears fast. A host overloaded by software decoding, transcodes or other containers can fall behind on I/O work. Diagnose disk latency, CPU load and network performance together instead of replacing a drive based on one symptom.

Plan exports separately from rolling retention

Rolling NVR storage is designed to overwrite old footage as retention windows expire. Important evidence should be exported to separate storage so it is not tied to the normal purge cycle. A second local disk, NAS backup or offsite copy can serve that role depending on sensitivity. Recording retention and evidence preservation are related but distinct storage problems.

Apply RAID overhead only after calculating footage

If the system needs mirroring, RAIDZ, RAID5/6 or another redundancy model, first calculate the video capacity required, then translate that into raw disk capacity. A 20 TB footage requirement does not mean a two-disk mirror provides 40 TB usable. Keep raw and usable numbers separate so the buying plan does not accidentally double-count or ignore redundancy overhead.

Recalculate with thirty days of real data

After deployment, record actual daily growth for several weeks. Compare measured data with the planned bitrate and duty cycle. Night noise, firmware updates, audio settings and scene complexity can change bitrate. Real usage lets you tune retention before capacity becomes critical and gives you a better forecast for expansion than any one-time calculator can provide.

Questions people ask

Frigate Recording Storage Guide FAQs

How is Frigate recording storage calculated?

Use total average recording bitrate, multiply by the amount of time footage is retained each day and by retention days, then add free-space headroom.

Does 4K always need more Frigate storage than 1080p?

Usually it can, but bitrate is the decisive input. Compression settings and scene complexity can make two cameras at the same resolution use very different storage.

Does Frigate re-encode recorded video?

Frigate says normal recordings are written directly from the camera stream without re-encoding, so the camera record bitrate drives storage growth.

How much headroom should I add?

A planning margin of roughly 15 to 25 percent is a practical starting point, with extra capacity added explicitly for future cameras.

Can I store Frigate recordings on NFS?

Yes, but test stability and copy performance. Frigate troubleshooting notes that network share protocol and mount behavior can affect recording performance.

Is SMB or NFS better for Frigate?

There is no universal answer, but Frigate troubleshooting notes that some users improved copy times with NFS. Test the exact NAS, network and mount options.

Should I use continuous or event recording?

Choose based on evidence needs. Continuous recording gives predictable maximum storage use; event-focused retention can save space but requires realistic activity assumptions.

How do I size RAID for Frigate?

Calculate the required usable footage capacity first, add headroom, then apply the chosen RAID or mirror efficiency to determine raw disk capacity.

What happens if Frigate storage is too slow?

Recording maintenance can fall behind, and troubleshooting should examine storage copy times, mount behavior, network stability and CPU load.

How often should retention be reviewed?

Review after several weeks of real operation and whenever cameras, bitrates or policies change. Actual daily growth should drive the next capacity decision.

Official references and methodology

Verify current Frigate support before deployment