Channelized Sub-interface

What Is a Channelized Sub-interface?

A channelized sub-interface is a sub-interface of an Ethernet physical interface with channelization enabled. Different channelized sub-interfaces are used to carry different types of services, and bandwidths are configured based on channelized sub-interfaces to implement strict bandwidth isolation between different channelized sub-interfaces on the same physical interface. This prevents services on different sub-interfaces from preempting each other's bandwidths. Channelized sub-interfaces are used to reserve resources in a network slicing solution. An independent "lane" is planned for each network slice, and "lanes" cannot be changed during transmission of different network slices' service traffic. This ensures strict isolation of services on different slices, and effectively prevents resource preemption between services when traffic bursts occur.

Why Do We Need a Channelized Sub-interface?

In the era of 5G and cloud services, diverse industries, services, and users present varying network service quality demands. Services such as mobile communication, environment monitoring, smart home, smart agriculture, and smart meter reading necessitate numerous device connections and frequent transmission of small packets. Conversely, live streaming, video uploading, mobile healthcare services, Internet of Vehicles (IoV), smart grid, and industrial control services demand higher transmission rates, millisecond-level latency, and near-100% reliability. These varied services entail distinct Quality of Service (QoS) requirements. To address them effectively, resource isolation for services is crucial. Initially, isolating services based on physical bandwidths sufficed. However, with the ongoing evolution of high-rate interfaces on routers, no service traffic can exclusively utilize high-rate interface bandwidth within a short timeframe. Consequently, high-rate interfaces are divided into low-bandwidth sub-interfaces to carry services, ensuring each sub-interface exclusively utilizes bandwidth resources and preventing bandwidth resource preemption. Channelized sub-interfaces facilitate this approach.

Channelized sub-interfaces serve to reserve resources within a network slicing solution, offering the following attributes:

• Strict resource isolation: Utilizing the sub-interface model, resources are pre-reserved to prevent slice services from preempting resources during traffic bursts.

• Small bandwidth granularity: Channelized sub-interfaces can be combined with FlexE interfaces, dividing high-rate interfaces into low-bandwidth sub-interfaces with a minimum granularity of 2M, ideal for industry slices.

What Are the Differences Between Channelized and Common Sub-interfaces?

The primary distinction between channelized and common sub-interfaces pertains to their ability to exclusively utilize bandwidth.

The remaining bandwidth of a main interface is calculated as the interface bandwidth minus the total bandwidth of all channelized sub-interfaces. For instance, upon configuring a 200 Mbit/s channelized sub-interface on a 1 Gbit/s interface, the main interface's bandwidth automatically decreases to 800 Mbit/s. A channelized sub-interface possesses exclusive bandwidth usage rights that cannot be overridden. Furthermore, resources can be earmarked for services to ensure service bandwidth. Conversely, the bandwidth of a common sub-interface can be overridden by other sub-interfaces, thus lacking stringent bandwidth assurance.

Channelized Sub-interface and FlexE Interface

What Is a FlexE Interface?

Flexible Ethernet (FlexE) technology separates the MAC layer from the PHY layer by introducing a FlexE shim layer (logical layer) between them. This shim layer enables the flexible mapping of MACs to PHYs, eliminating the need for a one-to-one mapping.

The PHY layer, which represents the physical Ethernet port, is organized into FlexE groups. The FlexE shim divides the PHY layer into adjustable timeslots, allowing for flexible allocation of bandwidth resources to FlexE clients, each occupying exclusive bandwidth. For example, two 100G physical interfaces are bonded to a FlexE group. If each timeslot is set to 5G, bandwidth resources can be flexibly allocated an integer multiple of 5G according to service requirements. For example, the FlexE group can be divided into three FlexE interfaces: 75G, 75G, and 50G. FlexE technology makes Ethernet interfaces "flexible."

FlexE structure

FlexE technology employs a FlexE shim to consolidate physical interface resources using timeslots. It dynamically segments a high-rate physical interface into multiple sub-channel interfaces (FlexE interfaces) using a timeslot resource pool, enabling adaptable and precise management of interface resources. Bandwidth resources of each FlexE interface are rigorously separated, ensuring that each operates akin to a physical interface, offering stringent assurance of bandwidth and latency.

What Are the Differences Between a Channelized Sub-interface and a FlexE Interface?

The key disparities between channelized sub-interfaces and FlexE interfaces are delineated as follows:

- Resource Isolation Effect

The methods for enforcing resource isolation differ between a FlexE interface and a channelized sub-interface. Illustrated in the diagram, FlexE interfaces achieve isolation based on timeslots between the MAC and PHY layers. Each FlexE interface possesses its own MAC layer, operating independently without interference from other FlexE interfaces during frame processing. Conversely, a channelized sub-interface lacks an independent MAC layer, relying on shared MAC layers. During frame processing, like jumbo frames, a channelized sub-interface waits until the current frame processing is complete before handling the next frame. Consequently, FlexE interfaces exhibit superior resource isolation.

Moreover, each channelized sub-interface features a distinct GQ/VI scheduling tree to enforce rigorous scheduling isolation, ensuring precise bandwidth and latency guarantees. Channelized sub-interfaces are suitable for carrying network services demanding assured bandwidth.

A FlexE interface mirrors a physical interface, boasting independent forwarding queues and buffers. FlexE interfaces experience minimal latency interference and deliver deterministic latency. They are ideal for carrying URLLC services demanding stringent latency SLAs, such as differential protection services for power grids.

Comparison between FlexE interfaces and channelized sub-interfaces

- Latency Assurance Effect

Under conditions of background traffic congestion, the maximum latency increase per hop differs significantly between channelized sub-interfaces and FlexE interfaces. While channelized sub-interfaces experience a maximum latency increase of 100 µs per hop, FlexE interfaces only incur a maximum increase of 10 µs per hop. Thus, FlexE interfaces offer lower latency performance, particularly crucial for latency-sensitive services.

Note: These specifications are provided based on the most severe congestion scenarios, in which the total interface traffic exceeds the interface bandwidth, and traffic congestion occurs in multiple reserved resources.

- Network Slice Granularity

Channelized sub-interfaces provide fine-grained slice granularity, with a minimum granularity of 2M. In contrast, FlexE interfaces offer relatively larger slice granularity, starting from 1G and supporting 1G, 1.25G, and 5G options.

How Does a Channelized Sub-interface Work?

What Is HQoS?

Understanding a channelized sub-interface entails grasping HQoS, as the two are interconnected. Unlike traditional QoS, which operates with single-level scheduling where a port differentiates services but not users, HQoS employs multi-level scheduling to refine traffic differentiation between users and services, enabling nuanced bandwidth management. This hierarchical QoS technology employs a multi-level queue scheduling mechanism within the DiffServ model to ensure multiple service bandwidth guarantees for various users.

Comparison between QoS and HQoS

HQoS facilitates differentiated quality services for diverse user and service traffic, unlike standard QoS. Nevertheless, heavy traffic on a main interface can lead to bandwidth preemption among sub-interfaces, impacting HQoS scheduling. Configuring a sub-interface to safeguard its allocated bandwidth from preemption ensures exclusive bandwidth utilization. Subsequently, scheduling isolation, based on the HQoS mechanism, is enforced for the sub-interface, rendering it a channelized sub-interface.

How Is a Channelized Sub-interface Implemented?

Channelized sub-interfaces adhere to the sub-interface model and leverage the HQoS mechanism to exclusively utilize the HQoS VI and GQ scheduling trees along with bandwidth to enforce stringent scheduling isolation. As shown in the following figure, when a 1 Gbit/s interface accommodates two channelized sub-interfaces (one at 200 Mbit/s and the other at 300 Mbit/s), the interface's bandwidth is reduced to 500 Mbit/s. This reduction occurs automatically upon enabling the channelized sub-interfaces, allowing them exclusive access to their allocated bandwidths. Each channelized sub-interface also possesses its own sub-interface scheduling tree, enabling independent queue scheduling. Consequently, various service types can be segregated by assigning different service traffic to distinct channelized sub-interfaces.

Implementation of channelized sub-interfaces

Channelized sub-interfaces offer management entity capabilities beyond HQoS. They combine the application scenario of a sub-interface with the bandwidth and management attributes of a main interface. On live networks, channelized sub-interfaces can work with a controller to manage resources, reserve resources in an E2E manner, and provide independent resources for a network slice.