News Details


How is the backplane bandwidth of an industrial switch calculated?

Release time:

2026-03-06

The backplane bandwidth of an industrial switch refers to the maximum amount of data that can be transferred between the switch’s CPU or interface cards and the system bus. It indicates the switch’s overall data‑transfer capacity, measured in Gbps, and is also known as the switching network bandwidth. Typically, industrial switches offer backplane bandwidth ranging from a few Gbps to several hundred Gbps. The higher a switch’s backplane bandwidth, the greater its ability to handle data traffic; however, this often comes with increased design and manufacturing costs. Next, we’ll take a closer look at the backplane bandwidth of industrial switches.

Generally speaking, the calculation method is as follows:

1) Wire-speed backplane bandwidth

Investigate the total network bandwidth available across all ports on the industrial switch. The calculation is: number of ports × port‑specific speed × 2 (for full‑duplex operation). If the total network bandwidth is less than or equal to the allowable backplane bandwidth, then the switch operates at line rate on the backplane.

2) Second-layer packet forwarding wire-speed

Layer 2 packet forwarding rate = (total number of Gigabit Ethernet ports × 1.488 Mpps) + (total number of 100-Mbps ports × 0.1488 Mpps) + (number of other port types × their respective calculation factors). If this forwarding rate is less than or equal to the allowable Layer 2 forwarding rate, then the industrial switch can maintain wire-speed performance when operating at Layer 2.

3) Third-layer packet forwarding wire-speed

Layer‑3 packet forwarding rate = (total number of Gigabit Ethernet ports × 1.488 Mpps) + (total number of 100‑Mb ports × 0.1488 Mpps) + (number of other port types × their respective calculation factors). If this rate is less than or equal to the allowable Layer‑3 forwarding rate, the industrial switch can maintain wire‑speed performance when operating at Layer 3.

So, how is 1.488 Mpps calculated?

The performance metric for wire‑speed packet forwarding is based on the number of 64‑byte data frames (the smallest possible frame) that can be forwarded per unit of time. For Gigabit Ethernet, the calculation is as follows: 1,000,000,000 bps ÷ 9 bytes ÷ (64 + 8 + 12) bytes = 1,488,095 pps. This indicates that when the Ethernet frame size is 64 bytes, one must account for the fixed overhead of a 9‑byte frame header and a 12‑byte interframe gap. Consequently, a wire‑speed Gigabit Ethernet port achieves a packet‑forwarding rate of 1.488 Mpps when handling 64‑byte packets. By contrast, the wire‑speed packet‑forwarding rate of a Fast Ethernet interface is exactly one‑tenth that of Gigabit Ethernet, or 148.9 kpps.

* For 10‑Gigabit Ethernet interfaces, the packet forwarding rate per wire-speed port is 14.88 Mpps.

* For Gigabit Ethernet, the packet forwarding rate of a wire-speed port is 1.488 Mpps.

* For a 100-Mbps Ethernet interface, the packet forwarding rate of a wire-speed port is 0.1488 Mpps.

* For the OC‑12 POS port, the packet forwarding rate of a wire-speed port is 1.17 Mpps.

* For OC‑48 POS ports, the packet forwarding rate of a line-rate port is 468 Mpps.

Therefore, if the three criteria mentioned above are taken into account, this industrial switch truly delivers line-rate, non-blocking performance.

The utilization rate of backplane bandwidth is closely related to the internal architecture of industrial switches. At present, the main types of internal architectures for industrial switches include the following:

First is the shared-memory architecture, which relies on a central management switching module to deliver high‑performance, full‑port connectivity. The core module examines each incoming packet to make forwarding decisions. This approach requires substantial memory bandwidth and incurs significant latency costs; moreover, as the number of ports in industrial switches grows, the cost of onboard memory rises sharply, making the switch’s core a bottleneck for performance.

Second, the cross‑connect structure can establish immediate point‑to‑point connections between port numbers; this works exceptionally well for multicast applications but is not well suited for multi‑point transmission.

Third is the hybrid crossbar architecture, a method for implementing a hybrid crossbar system bus. Its design principle involves partitioning the monolithic crossbar routing matrix into smaller crossbar routing sub‑matrices, with a high‑performance system bus connecting them at the center. The advantages include reducing the number of crossbar system buses, lowering costs, and mitigating bus contention; however, the system bus that interconnects these crossbar routing sub‑matrices introduces new characteristics.
Weakness.

The above content provides a detailed overview of industrial switch backplane bandwidth and is intended to be helpful to all.
 

Guangzhou Guangqiao Communication Equipment Co., Ltd. specializes in the R&D and manufacturing of communication equipment. Its main product lines include industrial switches, industrial Ethernet switches, industrial PoE switches, fiber optic transceivers, fiber optic switches, multi‑port transceivers, PoE switches, telephone optical terminals, video optical terminals, protocol converters (bridges), PCM multiplexing equipment, PDH optical terminals, E1 protection switching devices, serial‑port optical terminals, wireless bridges, SFP optical modules, and more. Contact numbers: 400‑8866‑640, 020‑82319301, 020‑82319302. Quality first, competitive pricing.

 

 

Previous page

Next page