Priority-based rate limits are used primarily by legacy devices. They are rate limits that are associated with one or more of the eight 802.1p priorities (0-7). When the associated priority is selected for a class of service, the rate limit becomes part of that class of service.
These rate limits are written directly to each port (unless the port is specified in the rate limit's exclusion list), and are implemented based on the 802.1p priority assigned to a data packet appearing on that port. While priority-based rate limits are not tied directly to roles or rules, they are displayed with the associated priority when you select a class of service while creating a rule, automated service, or role.
When priority-based rate limiting is implemented, the combined rate of all traffic on the port that matches the priorities associated with the rate limit cannot exceed the configured limit. If the rate exceeds the configured limit, frames are dropped until the rate falls below the limit.
Once a rate limit is associated with a priority, that priority will include rate limiting wherever and however it is used, until the rate limit is deleted from Policy Manager. Also, once a priority-based rate limit is applied to a port, it will remain on the port even if the role that originally used the rate limit is no longer associated with the port. For example, if an untagged packet arrives on a port where there is no role or default priority, but the port's 802.1p priority includes a rate limit, that traffic will be rate limited. As another example, if the priority of a tagged packet matches a priority-based rate limit on a port, the traffic will be rate limited.
To configure a priority-based rate limit, you need to specify the following components:
- Rate Limit - The highest transmission rate at which traffic can enter or exit a port.
- Direction - The direction to which the limit applies (inbound or outbound traffic). In order to control traffic inbound and outbound on the same port, two rate limits must be configured (one inbound and one outbound). Inbound rate limiting takes place after a frame has been classified into one of the eight priorities. Outbound rate limiting takes place just before a frame is queued for transmission. A single frame may pass through inbound and outbound rate limits depending on the path it takes through the device and the rate limiting configuration on the device.
- Priority - The 802.1p priority or priorities the rate limit is associated with.
- Precedence - The order in which the rate limit will be written to devices that support it. Policy Manager allows you to define as many rate limits as you wish; however, the number written to a device is restricted by the number of rate limits supported by the device. Each port on the device may utilize any or all of the defined rate limits up to the number of rate limits it supports.
- Exclusion - The devices/ports you wish to be excluded from the rate limit. For example, rate limiting is most often used for edge devices; therefore, you might want to exclude a device group or port group containing non-edge devices or ports.
