Page 9 | AlliedWare Plus™ OS: Overview of QoS
Scheduling
In addition to managing the way in which packets can be dropped when the egress queues for
a given port start to fill up, you can also configure the method that is used to allocate
bandwidth to each of the queues to transmit packets onto the line.
There are two ways that the queues can be scheduled for transmission:
zStrict Priority Scheduling
Higher-priority queues are emptied before any packets are transmitted from lower-
priority queues. This means that queue 7 must be totally empty before any packets from
queue 6 are transmitted, and so on.
zWeighted Round-Robin Scheduling
The queues share bandwidth on the basis of user-defined weights. Using this method,
packets from a lower-priority queue can be transmitted even when packets are waiting in
a higher-priority queue. The weights can be configured to ensure that more packets per
second are sent from the higher-priority queues than from the lower-priority queues.
To allow for flexibility in scheduling, it is possible to use different scheduling methods for
different queues. For a given port, you can create up to three groups of egress queues, one
that uses Strict Priority Scheduling and two separate groups that each use Weighted Round-
Robin Scheduling. For example, consider this case:
zqueues 7, 6 & 5 are configured to use Strict Priority Scheduling
zqueues 4, 3 & 2 are in Weighted Round-Robin group
1
zqueues
1
& 0 are in Weighted Round-Robin group 2
Queues 7, 6 & 5 will be emptied using Strict Priority, that is, queue 7 will be emptied before
any packets from queue 6 can be transmitted and queue 6 must be completely emptied
before any packets from queue 5 are transmitted.
When queues 7, 6 & 5 are all completely empty, queues 4, 3 & 2 will be emptied concurrently
based on their respective weights.
Queues
1
& 0 will be emptied only when there are no packets awaiting transmission in any of
the other queues.