Filter
Top Products
Configuring quality of service (QoS) 97
OnSite 2800 Series User Manual 8 • Link scheduler configuration
Each traffic-class is in fact assigned a relative weight, which is used to share the bandwidth among the currently
active classes. Patton recommends that you specify the weight as percent which is best readable.
Shaping
There is another commonly used way to assign bandwidth. It is called shaping and it makes sure that each traf-
fic-class will get just as much bandwidth as configured and not more. This is useful if you have subscribed to a
service that is only available for a limited bandwidth e.g. low delay. When connecting the OnSite to a DiffServ
network shaping might be a required operation.
Burst tolerant shaping or wfq
For weighted fair queuing and shaping there is a variation of the scheduler that allows to specify if a traffic class
may temporarily receive a higher rate as long as the average stays below the limit. This burstiness measure
allows the network to explicitly assign buffers to bursty sources.
When you use shaping on the access link the shaper sometimes has the problem that multiple sources are
scheduled for the same time - and therefore some of them will be served too late. If the rate of every source had
to strictly obey its limit, all following packets would also have to be delayed by the same amount, and further
collisions would reduce the achieved rate even further. To avoid this effect, the OnSite shaper assumes that the
burstiness needed for sources to catch up after collisions is implicitly allowed. Future versions of OnSite might
allow setting the burst rate and bursting size if more control over its behavior is considered necessary.
Burst tolerance has a different effect when used with weighted fair queuing. Think of it as a higher initial rate
when a source device starts transmitting data packets. This allows giving a higher weight to short data transfers.
This feature is sometimes referred to as a service curve.
Hierarchy
An arbiter can either use wfq or shaping to determine which source to serve next. If you want the scheduler to
follow a combination of decision criteria you can combine different schedulers in hierarchy to do a multi-level
arbitration.Hierarchical scheduling is supported in OnSite with service-policy profiles used inside service-pol-
icy profiles.In figure 19 an example of hierarchical scheduling is illustrated. The 1
st
level arbiter Level_1 uses
weighted fair queuing to share the bandwidth among source classes VPN, Web and incorporates the traffic
from the 2
nd
level arbiter Low_Priority, which itself uses shaping to share the bandwidth among source classes
Mail and Default.