How many surveillance cameras require a core switch?

How many surveillance cameras require a core switch? This is a familiar question security surveillance installation engineers often encounter when configuring security surveillance systems for clients. How many cameras typically require a core switch? Will the surveillance images lag if a core switch is not used during installation and configuration? Do I need a core switch for my 150-channel surveillance cameras? Many engineers also say that I can manage 300 cameras without a core switch, and that’s fine! With 10 years of experience as a security R&D engineer, I will tell you how to configure a core switch for cameras.

What is a core switch?

A network has three layers: access, aggregation, and core. A core switch, installed in the core layer, serves as the hub of the network architecture, primarily used for high-speed data exchange and connecting multiple subnets or LANs. Core switches generally offer high bandwidth, low latency, and high reliability, capable of handling large amounts of data traffic and ensuring efficient network operation.

How many surveillance cameras require a core switch?

For systems with fewer than 32 channels, a core switch is generally unnecessary. For systems with 32 or more cameras, the decision to use a core switch should be made based on the actual situation.

To determine whether a core switch is necessary, you must first understand the following key factors: the bandwidth (i.e., bitrate) consumed by each channel, the number of cameras, and the switch’s bandwidth capacity. Standard switches include 100M and 100M switches, but their actual bandwidth is typically only 60% to 70% of their theoretical bandwidth. Therefore, the available bandwidth per port is approximately 60 Mbps (100M switches) or 600 Mbps (100M switches).

The relationship between the camera’s bit rate and the switch’s bandwidth

For example, with a Hikvision 1.3-megapixel camera (960P), the bitrate is approximately 4 Mbps (H.264 encoding).

• Using a 100 Mbps switch, 15 cameras can be connected (15 × 4 = 60 Mbps).
• Using a Gigabit switch (1000 Mbps), 150 cameras can be connected (150 × 4 = 600 Mbps).

For example, with a Hikvision 2-megapixel camera (1080P), the bitrate per camera is approximately 8 Mbps (H.264 encoding).

• Using a 100 Mbps switch, you can connect seven cameras (7 × 8 = 56 Mbps).
• Using a 1000 Mbps switch, you can connect 75 cameras (75 × 8 = 600 Mbps).

These calculations assume cameras use H.264 compression. If you use H.265 compression, the bitrate will be reduced by about half.

For example, with a Hikvision 4-megapixel camera (1440P, H.265 encoding), the bitrate for each camera is approximately 4 Mbps (H.265 encoding).

In this case, the cameras’ bandwidth requirements are reduced, and with a Gigabit switch (1000 Mbps), 150 cameras can be connected (150 × 4 = 600 Mbps).

How to choose a switch? Detailed analysis of the selection criteria for access layer, aggregation layer, and core layer switches.

In large-scale video surveillance systems, switch selection is crucial. Switches must not only meet the system’s bandwidth requirements but also ensure network stability and scalability. Switch selection criteria vary depending on the functional requirements of different layers. This article will help you understand how to choose the right switch across three layers: the access layer, the aggregation layer, and the core layer.

Access layer switch selection

The access layer switch primarily connects surveillance cameras and transmits data to the aggregation layer. Taking 500 cameras as an example, assuming each camera has a 4 Mbps data rate, we need to analyze the following conditions:

Camera stream

Assume each camera has a bit rate of 4 Mbps. Based on this, we can calculate the bandwidth requirements of the access-layer switch.

The bandwidth required for 10 cameras is: 10 × 4 Mbps = 40 Mbps.

This means that the uplink ports of the access-layer switches must meet a transmission rate of at least 40 Mbps.

The actual bandwidth of the switch

There’s a discrepancy between the theoretical bandwidth of most switches and their actual available bandwidth. Typically, a switch’s actual bandwidth is 60%- 70% of its nominal value. For example, a 100 Mbps switch has an actual bandwidth of approximately 60 Mbps, while a 1000 Mbps switch has an actual bandwidth of 600 Mbps.

Using a 100 Mbps switch, each switch can connect to 10 cameras (10 × 4 Mbps = 40 Mbps).

The number of switches required for 500 cameras is: 500 ÷ 10 = 50 100M switches.

If a gigabit switch is used, each switch can connect to more cameras (up to 150 cameras). The specific choice depends on the system scale and budget.

Switch backplane bandwidth

The backplane bandwidth of each switch affects its data-forwarding capability. For example, the backplane bandwidth requirement for a 16-port 100M switch is:

16 × 100 Mbps × 2 = 3.2 Gbps (assuming bidirectional traffic).

Packet forwarding rate

The packet forwarding rate (PPS) of a switch determines its ability to process data packets. For example, the packet forwarding rate of a 1000M port is 1.488Mpps/s. If a 16-port 100M switch is used, the packet forwarding capacity is: (16 × 100M / 100M) × 0.1488 = 2.368Mpps.

Therefore, for a network with 500 cameras, you need 50 100M switches with a backplane bandwidth greater than 3.2 Gbps and a packet forwarding rate greater than 2.368 Mpps.

Selection of aggregation layer switches

Aggregation switches receive traffic from access switches, aggregate traffic from multiple access switches, and forward it to the core layer. The aggregation layer is under the greatest pressure, and the following factors must be considered:

Bandwidth requirements for aggregation layer switches

Assume there are 500 cameras, each with a 4 Mbps bitrate (H.264 encoding). If there are three switches at the aggregation layer, each aggregation switch needs to process:

Bandwidth requirement for 170 cameras: 170 × 4 Mbps = 680 Mbps.

Therefore, each switch in the aggregation layer must have a forwarding capacity of at least 680 Mbps. Generally speaking, a Gigabit switch (1000 Mbps) can barely meet this requirement, but if your budget allows, you can choose more switches to ensure redundancy and stability.

Bandwidth for storage and real-time viewing

Video storage is typically placed at the aggregation layer, and video recordings are forwarded at line speed. However, the bandwidth pressure of viewing surveillance videos in real time must be considered. Assuming each user views 16 channels of video simultaneously, each video stream requires 4 Mbps, and 10 users view simultaneously:

10 people × 16 channels × 4 Mbps = 640 Mbps.

In this case, the aggregation layer switch needs to have a bandwidth of at least 640 Mbps.

Core layer switch selection

The core layer switch is the core of the entire monitoring network. It is responsible for forwarding data from the aggregation layer switch to other parts. It does not have the pressure of video recording and is primarily responsible for transmitting real-time video streams. The selection of core switches should consider the following factors:

Switching capacity requirements

The bandwidth that the core switch needs to support depends mainly on the number of users watching simultaneously. For example, if 10 people are watching the system simultaneously, and each person watches 16 channels of video, the switching capacity requirement is:

10 × 16 × 4 Mbps = 640 Mbps.

Therefore, the core switch must meet a switching capacity requirement of at least 640 Mbps to support real-time viewing of multiple video streams.

Link bandwidth requirements

The link between the core switch and the aggregation layer must have sufficient bandwidth to support the forwarding of large amounts of data. If the storage and recording systems are located in the aggregation layer, the core switch’s bandwidth requirements depend on the link load from the aggregation to the core layers.

Conclusion

Depending on the number of cameras, bitrate requirements, and network layer, several factors should be considered when selecting a switch:

• Access-layer switch: Select an appropriate switch based on the camera’s bitrate requirements and bandwidth limitations. Either 100 Mbps or 100 Mbps switches are acceptable, depending on the network scale.
• Aggregation layer switch: Carries traffic from multiple access layer switches and handles the bandwidth requirements for live viewing and recording. Typically, a 100Mbps switch is sufficient, but additional redundancy can improve network reliability.
• Core layer switch: Primary responsibility for forwarding live video streams. Ensure sufficient switching capacity to support concurrent video viewing users.