Troubleshooting
Because many nodes might be attempting to join or rejoin the network after an expected or unexpected event,
a hold-off time of 16 minutes is implemented, which means that no nodes are exclusion-listed during this
period of time after system initialization.
This exponential backoff and advance algorithm is unique and has the following properties:
• It allows a node to correctly identify the parent nodes whether it is a true honeypot or is just experiencing
• It credits the good parent nodes according to the time it has enabled a node to stay connected with the
• It has a built-in hysteresis for encountering the initial condition issue where many nodes try to discover
• It has a built-in memory for nodes that can appear as neighbors sporadically so they are not accidentally
The node exclusion-listing algorithm guards the mesh network against serious stranding. It integrates into
AWPP in such a way that a node can quickly reconverge and find the correct network.
Throughput Analysis
Throughput depends on packet error rate and hop count.
Capacity and throughput are orthogonal concepts. Throughput is one user's experience at node N and the total
area capacity is calculated over the entire sector of N-nodes and is based on the number of ingress and egress
RAP, assuming separate noninterfering channels.
For example, 4 RAPs at 10 Mbps each deliver 40 Mbps total capacity. So, one user at 2 hops out, logically
under each RAP, could get 5 Mbps each of TPUT, but consume 40 Mbps of the backhaul capacity.
With the Cisco Mesh solution, the per-hop latency is less than 10 msecs, and the typical latency numbers per
hop range from 1 to 3 msecs. Overall jitter is also less than 3 msecs.
Throughput depends on the type of traffic being passed through the network: User Datagram Protocol (UDP)
or Transmission Control Protocol (TCP). UDP sends a packet over Ethernet with a source and destination
address and a UDP protocol header. It does not expect an acknowledgement (ACK). There is no assurance
that the packet is delivered at the application layer.
TCP is similar to UDP but it is a reliable packet delivery mechanism. There are packet acknowledgments and
a sliding window technique is used to allow the sender to transmit multiple packets before waiting for an
ACK. There is a maximum amount of data the client transmits (called a TCP socket buffer window) before
it stops sending data. Sequence numbers track packets sent and ensure that they arrive in the correct order.
TCP uses cumulative ACKs and the receiver reports how much of the current stream has been received. An
ACK might cover any number of packets, up to the TCP window size.
TCP uses slow start and multiplicative decrease to respond to network congestion or packet loss. When a
packet is lost, the TCP window is cut in half and the back-off retransmission timer is increased exponentially.
Wireless is subject to packet loss due to interference issues and TCP reacts to this packet loss. A slow start
recovery algorithm is also used to avoid swamping a connection when recovering from packet loss. The effect
of these algorithms in a lossy network environment is to lessen the overall throughput of a traffic stream.
By default, the maximum segment size (MSS) of TCP is 1460 bytes, which results in a 1500-byte IP datagram.
TCP fragments any data packet that is larger than 1460 bytes, which can cause at least a 30-percent throughput
drop. In addition, the controller encapsulates IP datagrams in the 48-byte CAPWAP tunnel header as shown
OL-27593-01
temporary outage conditions.
network. The crediting requires less and less time to bring the exclusion-list conviction period to be very
low for real transient conditions and not so low for transient to moderate outages.
each other only to find that those nodes are not really meant to be in the same network.
considered as parents if they were, or are supposed to be, on the exclusion-list database.
Cisco Mesh Access Points, Design and Deployment Guide, Release 7.3
Throughput Analysis
225