Go to documentation repository
| Section | |||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
...
For Axxon One VMS operation, the disk subsystem must match the following requirements:
...
This section provides the requirements for the performance (IOPS) and disk subsystem capacity for the correct operation of Axxon One.
The performance of the disk subsystem is measured in input/output operations per second (IOPS
...
). The disk (or array) must provide an IOPS value no lower than the estimated value, determined by the software load.
Axxon One interacts with the archive in fixed-size blocks:
...
Below you can see how you can estimate the IOPS for Axxon One VMS.
...
If the average bitrate per camera is less than ~3.6 Mbps
...
If the average bitrate per camera exceeds ~3.6 Mbps
...
IOPS during archive recording
Archive recording includes both data input (recording) and output (reading) operations
...
IOPS during archive playback
Archive playback includes only output (reading) operations
...
IOPS (reading) = 0.035 * R * S
...
IOPS during simultaneous recording and playback
...
where
...
The disk subsystem capacity of Axxon One Server must be estimated taking into account the resolution, the compression algorithm, the frame rate of the recorded video signal, the number of video cameras from which the recording to disk(s) takes place and other recording parameters. In addition, the size of the system log and metadata databases must be considered.
You need at least 10 GB of free disk space on Windows OS and 5 GB on Linux OS to install the Axxon One VMS package in the Server and Client configuration.
See the storage calculations below, taking into account the size of the archive, the size of the database and the size of the archive of the system logs.
The capacity of the disk subsystem can be calculated by the formula:
Capacity of disk subsystem (MB) = Time of storing archive (days) * Cameras number * Rate of recording (FPS) * 3.51 * Time of guaranteed recording from a camera (h/day) * Average frame size (KB),
where
Time of storing archive is the required time for storing an archive from one camera, days;
Cameras number is the number of cameras from which recording to the archive takes place;
Rate of recording is the frame rate of recording to the archive, frames per second;
3.51 = (60 sec in min * 60 min in hour) / (1024 KB in MB) is the factor used for KB/s in MB/h conversion;
Time of guaranteed recording from a camera is the number of hours of guaranteed recording from one camera per day;
Average frame size is the average size of the camera frame in KB.
| Info | ||||||||
|---|---|---|---|---|---|---|---|---|
| ||||||||
Average frame size of 640х480 resolution is:
Average frame size may vary over a wide range depending on the vendor, the model and the settings of the camera and video image complexity. |
| Info | ||
|---|---|---|
| ||
To calculate the frame size you can use the ratio, that when increasing the vertical or horizontal resolution two times, the average frame size will be increased four times (this rule is a relative and can be applied only to some camera models). |
Examples of calculating the capacity of the disk subsystem (without the capacity of the system log and metadata databases) are presented below.
...
Recording parameters
...
Calculating results
...
4 cameras with 25 FPS and 640х480 resolution, guaranteed recording of 24 hours per day during one week
...
H.264: from 500 GB to 1 TB
MPEG4: from 500 GB to 2 TB
MJPEG: from 1.3 TB to 3.5 TB
...
16 cameras with 12 FPS and 640х480 resolution, guaranteed recording of 12 hours per day during one week
...
H.264: from 500 GB to 1 TB
MPEG4: from 500 GB to 2 TB
MJPEG: from 1.3 TB to 3.5 TB
...
4 cameras with 25 FPS and 1280х960 resolution, guaranteed recording of 24 hours per day during one week
...
H.264: from 2 TB to 4 TB
MPEG4: from 2 TB to 8 TB
MJPEG: from 5.3 TB to 14 TB
Below are the formulas for calculating IOPS depending on the usage case and the average bitrate of the cameras.
| Note | ||
|---|---|---|
| ||
The 3.6 Mbps limit is approximate. When calculating, keep in mind that when writing an archive, both write (input) and read (output) operations are performed simultaneously, which are necessary for service purposes. |
| Camera average bitrate | Write IOPS | Read IOPS | ||
|---|---|---|---|---|
| Archive recording only | ||||
| Less than ~3.6 Mbps | 0.29 × N | 0.035 × M | ||
| More than ~3.6 Mbps | 0.065 × M | 0.035 × M | ||
| Archive playback only | ||||
| When viewing, only read operations are performed | 0.035 × R × S | |||
| Simultaneous recording and viewing of the archive | ||||
| Less than ~3.6 Mbps | 0.29 × N | 0.035 × M + 0.035 × R × S | ||
| More than ~3.6 Mbps | 0.065 × M | 0.035 × M + 0.035 × R × S | ||
| ||||
When you use RAID controllers, you must set the cache policy to Write-Back to ensure maximum write performance.
The total disk space consists of the capacity required to store video data (archive), system log, metadata, and service information.
To install the Axxon One distribution package, you need:
| Note | ||
|---|---|---|
| ||
The specified values don't include the space for the video archive (see Аrchive), metadata database (see Metadata database), system log (see System log), and DetectorPack add-on (see Installing DetectorPack add-ons). The calculations for them are provided below. |
The estimated disk space required to store a video archive can be determined using the following formula:
Capacity (MB) = Storage T x N cameras x Bitrate x 450 x Recording T
| Tip |
|---|
Names of variables
|
The average bitrate depends on the codec, camera manufacturer, compression settings, scene dynamics, resolution, and hardware settings (see Platform Calculator). Examples for 1920x1080 resolution:
| Video codec | Bitrate |
|---|---|
| H.264 | 4 Mbps |
| H.265 | 2.6 Mbps |
| Recording parameters | Calculation results (total capacity) |
|---|---|
| 4 cameras, 25 FPS, 1920×1080, 24/7 recording, 7-day storage |
|
The size of the system log database depends on the number of configured detectors and their activity.
| Detector activity | Сalculation formula | ||
|---|---|---|---|
| Low | D × T × 0.04 GB | ||
| Medium | D × T × 0.12 GB | ||
| High | D × T × 0.48 GB | ||
| |||
We recommend calculating the size for the metadata database using one of the formulas, depending on the need to reserve space.
| Size | Сalculation formula | |||||||
|---|---|---|---|---|---|---|---|---|
| Sufficient | N × T × 0.5 GB | |||||||
| With margin | N × T × 1 GB | |||||||
| With a large margin | N × T × 5 GB | |||||||
| ||||||||
The capacity of the system log database must be taken into account when the capacity of the disk subsystem is calculated. The estimated capacity of the system log database is calculated by the formulas:
Capacity of system log database (low detection tools activity) = D * T * (0.04 GB / day);
Capacity of system log database (average detection tools activity) = D * T * (0.12 GB / day);
Capacity of system log database (high detection tools activity) = D * T * (0.48 GB / day);
where
D is the total number of detection tools created in the system,
T is the estimated duration of system log storage, days.
The following formulas can be used to determine the required disk size for the metadata database:
Size of metadata database = N * T * (0.5 GB / day) – sufficient disk size;
Size of metadata database = N * T * (1 GB / day) – sufficient disk size plus reserve space;
Size of metadata database = N * T * (5 GB / day) – sufficient disk size plus a large reserve,
where
N is the number of detection tools in the system actively recording metadata. One video camera can have several detection tools with metadata (see Metadata database),
T is the period of time (number of days) that metadata will be stored. By default, T = 30 days.
...
| ||
...
When calculating the disk subsystem capacity, it is necessary to include you plan disk space, consider the following:
...
...
...
...
...
...