Lead Forensics

What is the difference between block storage and file storage for CCTV archives?

Block storage presents raw volumes to a server, which formats and manages them with its own file system, while file storage presents a ready-made shared file system over a network. For CCTV archives, block storage (typically SAN) suits high-throughput, sustained video recording, whereas file storage (typically NAS) suits simpler, shared or lower-intensity retention. The right choice depends on bitrate, camera count, VMS requirements and resilience targets, not on a single “better” architecture. The sections below cover how each handles continuous writes, performance, VMS compatibility, resilience and sizing for surveillance storage projects.

How does each storage type handle continuous video writes?

Block storage handles continuous video writes by exposing raw volumes directly to the recording server, which writes sequential streams with minimal protocol overhead. File storage routes every write through a network file protocol such as SMB or NFS, adding processing layers between the recording server and the disks.

CCTV recording is a sustained, mostly sequential write workload. Dozens or hundreds of camera streams arrive around the clock, and unlike a database, the pattern is predictable and continuous. The question is how efficiently the storage path carries that load without stalls.

With block storage, the recording server owns the file system and controls how video is laid out on the volume. Many VMS platforms tune their archive writing around this assumption. With file storage, the NAS device owns the file system, and the recording server hands each write over a network protocol. That can work well, but file locking, protocol latency and network contention become additional variables.

Neither approach is automatically fragile. The issue is headroom: the storage path must sustain the aggregate write rate during normal operation and while a drive is rebuilding.

Which storage type performs better for HD surveillance recording?

Block storage generally performs better for large HD surveillance recording because it offers lower latency and more predictable sustained write throughput. File storage can perform adequately for smaller or moderate deployments, but protocol overhead and shared network paths make performance harder to guarantee as camera counts and bitrates rise.

Performance in surveillance storage is about consistency rather than peak speed. A single dropped write can mean a lost frame, so the measure that matters is whether throughput holds steady under the full recording load, including analytics activity and playback requests at the same time.

Key factors that influence the outcome include:

  • Aggregate bitrate: the total of all camera streams, including any headroom for increased frame rates or resolution upgrades.
  • Network design: dedicated storage networks protect recording traffic from competing with general data. This matters for both approaches.
  • Drive and RAID configuration: the array layout, controller and cache behaviour often limit throughput more than the access method does.
  • Concurrent playback and analytics: reads and metadata writes compete with recording writes unless the design accounts for them.

Treat any vendor throughput figure as indicative until it is validated against your actual camera mix and VMS configuration. Secure Logiq’s server selection guidance covers the factors that shape that validation.

Do VMS platforms work better with block or file storage?

Most VMS platforms work most dependably with block storage because they treat the archive volume as a local disk, giving them direct control over file handling. File storage is often supported, but some platforms restrict it or recommend specific configurations, so compatibility must be confirmed against the vendor’s current documentation.

This is where many designs go wrong. A storage architecture that looks sound on paper can fall outside what a VMS vendor supports for archive recording. Common concerns include how the VMS handles network share permissions, behaviour when a network path drops momentarily, and whether recording continues cleanly when the share reconnects.

Before specifying either type, check three things with the VMS manufacturer’s documentation:

  1. Whether network file shares are supported for live recording archives, or only for secondary or archived storage tiers.
  2. Any recommended protocol versions, authentication methods or mount configurations.
  3. Whether the platform has specific requirements for failover recording servers sharing the same archive.

Block storage presented to the recording server as a local volume sidesteps many of these questions, which is one reason it is common in enterprise installations. It is still worth verifying, as support statements change between VMS releases.

When should a CCTV archive use file storage instead?

A CCTV archive should use file storage when the deployment is small to moderate, when footage must be shared across multiple servers or users, or when the VMS explicitly supports network shares for the archive tier. It also suits secondary retention, such as longer-term copies of exported or older footage.

File storage has real advantages that make it the sensible choice in the right circumstances:

  • Simplicity: NAS devices are straightforward to deploy and manage, which suits smaller sites without dedicated storage administrators.
  • Shared access: multiple clients or servers can reach the same file system, which helps with evidence export, review workflows and secondary archives.
  • Tiered retention: older footage that no longer needs high write performance can sit on cheaper, simpler file storage.
  • Lower entry cost: for modest camera counts, a well-specified NAS can be more cost-effective than a full SAN.

The limitation is that performance headroom and VMS support become the deciding constraints as the system grows. A file storage design that works for fifty cameras may need rethinking at several hundred, particularly with analytics-heavy workloads.

How do resilience and scalability differ between the two?

Both types can be made resilient through RAID, redundant controllers and dual power, but they scale differently. Block storage scales by adding capacity and bandwidth to the storage network, often behind multiple recording servers. File storage scales by expanding the NAS or adding nodes, with the file protocol layer remaining a shared dependency.

Resilience in either case needs to be designed deliberately, and it is worth remembering the following distinctions:

  • RAID is not a backup. It protects against drive failure, not deletion, corruption or site loss.
  • Redundancy does not guarantee uninterrupted recording. Rebuild performance, controller failover behaviour and network path redundancy all affect whether recording continues cleanly during a fault.
  • Single points of failure move rather than disappear. A resilient array behind a single switch or a single network link is still exposed.

On scalability, SAN architectures typically allow capacity and recording servers to be expanded independently, which suits large estates and multi-site consolidation. NAS expansion is simpler at small scale but can hit throughput ceilings at the file service layer. Secure Logiq’s Qatar University SAN storage case study shows how a SAN approach was applied to a large campus requirement.

How should storage be sized for a CCTV archive?

Size a CCTV archive by multiplying total bitrate by retention time, then adding allowances for RAID overhead, spare capacity, metadata and future growth. The result is the raw capacity needed, but throughput, rebuild behaviour and VMS limits must also be checked, because capacity alone does not guarantee reliable recording.

A basic method, explained with its assumptions, looks like this. The figures below are a hypothetical illustration only, not a performance claim or recommendation:

  1. Calculate aggregate bitrate. For example, 200 cameras at an assumed average of 4 Mbps gives 800 Mbps of continuous recording.
  2. Convert to daily storage. 800 Mbps is 100 MB per second, which is roughly 8.6 TB per day (100 MB/s multiplied by 86,400 seconds).
  3. Multiply by retention. For 30 days, that is approximately 259 TB of usable video capacity.
  4. Add overheads. Account for RAID parity or mirroring, hot spares, file system overhead, analytics metadata and a growth margin.

Real bitrates vary with codec, scene complexity, frame rate and motion recording settings, so use measured or manufacturer-tested figures wherever possible. Also confirm that the design sustains the write rate during a rebuild and alongside playback.

Because capacity, throughput and VMS support all interact, sizing is best treated as a system design exercise rather than a spreadsheet exercise. If you are planning a new archive or reviewing an existing one, Secure Logiq’s system design service can help you validate block or file storage choices against your actual camera estate and retention targets.

Frequently Asked Questions

How do I get started choosing between block and file storage for a new CCTV project?

Start by documenting your camera count, average and peak bitrates, retention period and VMS platform. Then check the VMS vendor’s current documentation to see whether network file shares are supported for live recording archives. Shortlist block or file storage based on those constraints, and validate the design with a pilot or a vendor-tested configuration before committing to a full purchase.

What happens to recording if a drive fails and the array starts rebuilding?

During a rebuild, the array is doing extra reads and writes to restore redundancy, which can reduce the throughput available for recording. If the design has no spare headroom, this can cause dropped frames or gaps in footage. Size the array so it can sustain the full aggregate write rate while rebuilding, and ask vendors for rebuild-time performance figures, not just normal-operation numbers.

Can I mix block and file storage in the same surveillance system?

Yes, and it is a common approach. Block storage (such as a SAN) is typically used for the primary recording tier where sustained write performance matters most, while file storage (such as a NAS) holds older footage, exported evidence or secondary copies. Confirm that your VMS supports tiered or secondary archive locations and how it moves footage between them.

What are the most common mistakes when sizing a CCTV archive?

Frequent errors include using nominal camera bitrates instead of measured ones, forgetting RAID overhead, hot spares and metadata, and leaving no margin for future cameras or higher resolutions. Another is sizing for capacity alone and ignoring throughput during playback, analytics and rebuilds. Build in a growth margin and test against realistic scene complexity and motion settings.

Do I need a dedicated network for surveillance storage?

A dedicated or well-segregated storage network is strongly recommended for both block and file approaches. Recording traffic is continuous, so competing with general office or data traffic can cause latency spikes and lost frames. At minimum, use separate VLANs with appropriate quality of service, and add redundant links and switches where uptime targets require it.

Is RAID enough to protect my footage?

No. RAID protects against drive failure but not against accidental deletion, file system corruption, ransomware or loss of the whole site. If footage has evidential or compliance value, add a separate protection layer such as a secondary archive, offsite copy or immutable retention, and test restores periodically so you know they actually work.

How do I validate a vendor's throughput claims before buying?

Treat published figures as indicative and ask for testing that reflects your own camera mix, codec, VMS and retention settings. Where possible, run a proof of concept that records at full aggregate bitrate while also performing playback, analytics and a simulated drive failure. Monitor for dropped frames, latency and write stalls, and only sign off the design if it holds steady across all those conditions.

Related Articles