CCTV Storage Calculator: Formula, Bitrate Table and HDD Recording Days

A CCTV storage calculator comes down to one rule: multiply each camera's bitrate in Mbps by 10.8 to get gigabytes per day, then multiply by the number of cameras and the retention days. One camera at 4 Mbps recording around the clock uses about 43 GB a day, so eight cameras kept for 30 days need roughly 10.4 TB of usable space.

This guide works as a CCTV storage calculator you can check by hand: the formula, quick reference tables, the reverse calculation for an existing hard disk, RAID and headroom, retention rules in India, and worked examples for real site types.

Key takeaways

• Storage (GB) = Bitrate (Mbps) × 10.8 × Cameras × Days, for continuous 24-hour recording.

• Bitrate is the number most often wrong. Measure it on site instead of trusting a datasheet, especially at night.

• H.265 typically needs about half the bitrate of H.264 at the same resolution, with the biggest savings on quiet scenes.

• Usable capacity is always lower than the label: allow for decimal versus binary units, formatting, RAID parity and 10 to 20% headroom.

• Size storage against the longest retention obligation that applies to the site, not the average.

What decides how much CCTV storage you need?

Most storage shortfalls are not caused by buying the wrong hard disk. They come from guessing one number: bitrate. Everything else in the calculation is arithmetic you can verify in a minute, but bitrate changes with resolution, compression, frame rate, lighting and how busy the scene is. Get it wrong by a factor of two, and 30-day retention quietly becomes 15 days.

Five inputs decide the answer:

1. Bitrate per camera in megabits per second (Mbps): the rate at which each camera writes video to disk, not the sensor resolution.

2. Number of cameras recording to that recorder or array.

3. Recording hours per day: 24 for continuous recording, fewer if cameras record only during operating hours.

4. Retention days: how far back you must be able to look.

5. Recording mode: continuous, motion-triggered, or event-triggered with a pre- and post-event buffer.

Bitrate is where errors cluster. Datasheets quote a maximum, installers quote a typical figure, and the real number drifts upward at night when infrared illumination adds sensor noise. A dome camera watching a still corridor and the same model watching a loading bay will not produce the same file sizes, even with identical settings.

Analytics change the picture again. A camera with onboard AI may record a second, lower-resolution substream for analysis, or send metadata alongside video. Edge AI reduces the bandwidth sent upstream but does not reduce what the recorder writes unless you configure it to. On multi-site projects the recorders and disks often become a larger line item than the cameras, so size them from measured bitrate, not a brochure. For what AI adds at the camera, see our guide to AI CCTV cameras in India.

CCTV storage calculation formula, with a worked example

This formula converts a per-second data rate into total capacity. Bitrate is in megabits and storage is sold in gigabytes, so dividing by 8 converts bits to bytes and dividing by 1,000 converts megabytes to gigabytes.

Storage (GB) = Bitrate (Mbps) × 3,600 × Hours per day × Days × Cameras ÷ 8 ÷ 1,000

For continuous recording this simplifies to Bitrate × 10.8 × Cameras × Days. Work through it in this order, checking each number before moving on:

1. Confirm the real bitrate. Record one camera for an hour at production settings during the busiest part of the day, then check the file size. Divide the megabytes by 450 to get approximate Mbps. If it differs sharply from the datasheet, trust your measurement.

2. Megabits per hour: multiply bitrate by 3,600. A 4 Mbps stream produces 14,400 megabits an hour.

3. Per day: multiply by recording hours. For continuous recording, 14,400 × 24 = 345,600 megabits per camera per day.

4. Retention: for 30 days, 345,600 × 30 = 10,368,000 megabits per camera.

5. Cameras: for eight cameras, 10,368,000 × 8 = 82,944,000 megabits.

6. Convert: ÷ 8 = 10,368,000 MB, then ÷ 1,000 = 10,368 GB, or about 10.4 TB.

Worked example, a warehouse with eight cameras: eight 4 MP cameras on H.265 at a measured 4 Mbps, recording continuously for 30 days, need about 10.4 TB of usable capacity. That is usable space after formatting and RAID parity, not raw disk capacity.

Put the same arithmetic in a spreadsheet and you have your own storage calculator, more reliable than a vendor calculator that assumes bitrates you have not verified. Change one input at a time: halving retention halves the total, and so does moving from continuous to 12-hour recording.

Typical CCTV bitrate by resolution: H.264 vs H.265

H.265 (HEVC) encodes the same scene at a lower bitrate than H.264 by describing large blocks of unchanged pixels more efficiently. The gain is real but not fixed: a car park at 3 am compresses to almost nothing, while a busy platform in the rain narrows the gap considerably.

Treat these as planning ranges, not guarantees. Four factors push a camera toward the top of its band:

• Frame rate: doubling from 15 to 30 fps comes close to doubling bitrate on active scenes. Most evidential needs are met at 12 to 15 fps.

• Scene activity: foliage, flags, rain, conveyor belts and traffic force the encoder to send more change data. A camera facing a tree line costs more disk than one facing a wall.

• Night noise: infrared mode adds sensor grain the encoder treats as motion. Many systems record more data between midnight and dawn than during the day.

• Audio: a few hundred kilobits per second per channel looks trivial until you multiply by 200 cameras and 90 days.

Before relying on any bitrate calculator, check whether its figures assume constant or variable bitrate, and whether they reflect daytime conditions only. Our guide to CCTV video compression techniques explains the codecs in more detail.

CCTV storage calculator: quick reference table

Use this table to estimate per-camera storage without a spreadsheet. All figures assume continuous 24-hour recording by one camera.

The one number worth memorising: 1 Mbps uses about 10.8 GB per camera per day. Multiply by your bitrate, camera count and retention days. A 40-camera site at 4 Mbps with 30-day retention needs about 51.8 TB of usable capacity.

The last column answers the question buyers ask most often. One terabyte of usable space holds about 23 days from a single 4 Mbps camera, or under three days if eight such cameras share it.

How many recording days will a CCTV hard disk give?

When the hardware is already fixed, reverse the formula to see what retention it supports:

Days = Usable capacity (GB) × 8,000 ÷ (Bitrate × 3,600 × Hours per day × Cameras)

A 16 TB usable array serving 24 cameras at 4 Mbps continuously gives 16,000 × 8,000 ÷ (4 × 3,600 × 24 × 24), or about 15.4 days. If policy requires 30 days, you are short by half, and no recorder setting will create the missing space.

Two things usually make the real answer worse than the calculated one:

• Decimal versus binary units: disks are sold in decimal terabytes (1 TB = 1012 bytes), but operating systems and many recorders report binary units (1 TiB = 1,099,511,627,776 bytes). A disk sold as 8 TB shows as about 7.27 TB once installed. Always calculate against reported usable capacity, not the label on the box.

• Overwrite behaviour: most NVRs run a circular buffer, deleting the oldest recordings when the volume fills. They do not warn you that retention has dropped. A site set for 30 days that quietly settles at 18 looks healthy until someone asks for footage from three weeks ago.

Check the oldest available recording on each recorder every month and compare it with your policy. That one check catches undersized arrays, failed disks and bitrate drift that no NVR storage calculator can predict. CheckCam camera health monitoring automates this by flagging recording gaps and storage faults across sites, and our guide to preventing CCTV recording downtime covers the manual checks.

Motion vs continuous recording: how much storage do you save?

Motion recording writes only when the detector fires, so storage scales with how much of the day the scene is active. Multiply your continuous figure by the activity percentage.

Two cautions before converting a site to motion recording:

• The pre-event buffer matters more than the trigger. A detector that starts recording when it fires has already missed the approach. Set a 5 to 10 second pre-roll and a similar post-roll. It adds back some storage but keeps the context that makes footage usable as evidence.

• Motion detection fails when you most need it. Rain, swinging lights, insects on the lens and moving shadows cause false triggers, while a slow intruder at the edge of the frame may cause none. False triggers push the camera toward continuous bitrate anyway.

Keep evidence-critical positions on continuous recording: cash handling points, entrances and exits, custody areas, data centre doors and cash-in-transit bays. Use motion recording for storerooms, corridors and interior coverage where the question is "was anyone here?" rather than "what exactly happened?".

RAID, overhead and headroom: sizing the NVR or storage server

Every calculation so far gives a usable capacity requirement. Raw disk capacity is a larger number, and the gap between the two is where sizing goes wrong. Three deductions apply before any video is written: decimal-to-binary conversion removes roughly 9%, filesystem formatting takes a further slice, and RAID parity reserves whole disks.

RAID 5 is common for surveillance because it costs only one disk of capacity. On large arrays of high-capacity disks, RAID 6 is safer: rebuilds take a long time, and a second failure during a rebuild is a real risk.

Then add headroom. Arrays that run at or near full capacity suffer fragmentation, slower writes and failed recordings at peak times, so plan 10 to 20% free space on top of the retention requirement. The 10.4 TB warehouse example therefore needs about 12 TB usable: on 8 TB disks in RAID 5, that is two data disks plus one parity disk, three in total, with a fourth giving room to grow.

NAS and NVR storage calculators often apply their own parity model, so check which deductions are already included before adding yours. Doubling a deduction damages a budget as much as leaving one out.

How long must CCTV footage be kept in India?

Retention is rarely a free choice. Regulators, licensing conditions and courts set minimum periods, and the obligation usually sits with the operator of the premises rather than the installer.

Schools:

CBSE requires affiliated schools to install high-resolution CCTV with audio-visual recording at entrances, classrooms, corridors and common areas (not toilets), with footage kept for at least 15 days (SCC Online, Jul 2025). See our education security solutions.

Police stations:

Public stations in Paramvir Singh Saini v Baljit Singh, the Supreme Court directed CCTV with night vision and audio in police stations, with footage preserved for 18 months and in no case less than one year (India Legal). These windows drive very large storage needs.

Banks and regulated businesses:

Banks, jewellers and other high-value sites typically keep footage for longer, often 60 to 90 days, driven by security policy and insurer requirements. Confirm the figure that applies to your institution in writing.

Court directions:

Courts can order footage preserved in specific matters, which overrides your routine overwrite cycle. Build a way to lock and export selected recordings before the circular buffer reaches them.

Two practical rules follow. Size storage against the longest obligation that applies to the site, and document which requirement drove your retention figure, because auditors and procurement reviewers will ask. The DPDP Rules, 2025 also limit how long personal data such as identifiable footage should be kept and who may access it, so over-retention carries its own risk.

Worked examples for Indian sites

The same formula applied to four site types shows how quickly the hardware changes.

• Retail store, 12 cameras: twelve 2 MP cameras on H.265 at 2 Mbps, continuous, 30-day retention: 12 × 2 × 10.8 × 30 = 7,776 GB, about 7.8 TB usable, or about 9 TB with 15% headroom. A RAID 5 array of three 6 TB disks covers it. See our retail security solutions.

• Manufacturing plant, 60 cameras: 45 cameras at 4 Mbps and 15 at 8 Mbps on gates and production lines, continuous, 45-day retention. The first group needs 45 × 4 × 10.8 × 45 = 87,480 GB and the second 15 × 8 × 10.8 × 45 = 58,320 GB: about 146 TB usable before headroom. See industrial and infrastructure solutions.

• Bank branch, 20 cameras: twenty cameras at 4 Mbps, continuous, with a 90-day retention policy: 20 × 4 × 10.8 × 90 = 77,760 GB, about 78 TB usable for one branch. Across a branch network, this number shapes the whole architecture. See banking and finance solutions.

• Police station, 25 cameras: twenty-five cameras at 4 Mbps, continuous. At the one-year minimum: 25 × 4 × 10.8 × 365 = 394,200 GB, about 394 TB usable per station. At the full 18 months, it is about 590 TB. See government and public sector solutions.

The last example is the one that breaks budgets. Long retention mandates are not met by adding disks indefinitely; they force decisions about compression, resolution tiers and archive storage. Check network load too: 25 streams at 4 Mbps is 100 Mbps of sustained traffic before any live viewing.

How to reduce CCTV storage without losing evidence

Lowering resolution is the one move that destroys evidential value, and it is the one most often made under budget pressure. These options are safer:

1. Move to H.265 where the hardware supports it, after confirming the recorder, client software and video management system can all decode it.

2. Set 12 to 15 fps on general coverage. Keep 25 to 30 fps for tills, counters and cash handling, where hand movement matters.

3. Use variable bitrate with a cap, so quiet scenes use less while a noisy night stream cannot run away with the array.

4. Apply motion recording selectively, with a proper pre-event buffer, keeping entrances, exits and evidence-critical positions continuous.

5. Tier retention: keep every camera at full retention for the first week, then only key cameras for the rest of the policy period, where regulations allow.

6. Fix noisy scenes physically: better lighting, repositioning away from moving foliage and clean lens housings all cut night-time bitrate.

For more practical steps, see our tips to reduce CCTV storage costs and CCTV footage storage best practices.

Keep more footage on the drives you already have

When these levers are exhausted and retention mandates remain, compression of recorded footage is the next step. CamStore by Transline Technologies compresses CCTV footage by up to 80%, so existing storage holds months more footage with no new hardware, while recordings stay available for playback and review.

Pair it with CamPulse, Transline's video management system with built-in camera health monitoring, so a silently failed camera never becomes the gap in a 90-day archive. Explore our wider video surveillance solutions.

Plan storage around your retention rules

Talk to our team about sizing storage for your sites and retention obligations.