Quality of Service is implemented in LCOS by means of different queues for the data packets. The following queues are used for data transmission:
- Urgent queue I
This queue is always processed first. The following data packets are handled here:
- Packets with ToS "Low Delay"
- Packets with DiffServ “Expedited Forwarding“
- All packets that have been assigned a certain minimum bandwidth as long as the guaranteed minimum bandwidth is not exceeded
- TCP control packets can also be sent preferentially through this queue
- Urgent queue II Packets that end up here have been assigned a guaranteed minimum bandwidth, but their connection has exceeded this bandwidth. Until the interval for the minimum bandwidth has expired (e.g. until the end of the current second), all packets in this queue are handled without any special priority. All of the packets in this queue, the "secured queue" and the "standard queue" now share the available bandwidth. Packets being sent are retrieved from the queues in the same order in which they were placed in the queues. If the interval expires, all blocks remaining in the Urgent Queue II are put back into the Urgent Queue I until the respectively assigned minimum bandwidth is exceeded, whereafter the rest remains in the Urgent Queue II. This procedure ensures that prioritized connections do not overwhelm the rest of the traffic.
- Secured queue
This queue has no separate priority. However, packets in this queue are never dropped (transmission guaranteed). The following data packets are handled here:
- Packets with ToS “High Reliability“
- Packets with DiffServ “Assured Forwarding"
- Standard queue
The standard queue contains all unclassified data packets. Packets in this queue are initially dropped if the data packets cannot be delivered fast enough.
The concept of queues only works if data packets are congested at the interface from the LAN to the WAN. Congestion occurs at peak times when data arrives from the LAN quicker than the device interface can deliver it to the WAN. This may be the case when the interface to the WAN is an integrated ADSL interface with a comparatively low upstream transmission speed. The integrated ADSL modem automatically reports back to the device how many data packets it can still accommodate, so that the data flow is slowed in the router. The queues will then be filled automatically.
Things look different when an Ethernet interface connects to the WAN. From the device's point of view, the connection to the Internet via an external DSL modem looks like an Ethernet segment. Between the device and the DSL modem, data is transmitted at the full LAN speed of 10, 100, 1000 or more Mbps. The input and output speeds are the same, so no natural congestion can arise here. Furthermore, the Ethernet between the device and the broadband modem does not report anything about the capacity of the connection. Consequently, congestion first arises in the DSL modem. Since no queues are available here, surplus data is lost. It is not possible to prioritize the preferred data.
To solve this problem, the transmission rate of the WAN interface in the device is reduced artificially. The interface is set to the transmission rate that is available for transporting data to the WAN.
Note: In most cases, the data rate specified by the providers is the net data rate. The gross data rate available from the interface is slightly higher than the net data rate guaranteed by the provider. If you know the gross data rate offered by your provider, you can enter this value for the interface and thus slightly increase the data throughput. By entering the net data rate you are certainly on the safe side.
