Nortel Networks NN43001-563 Network delay and packet loss evaluation example

146 ITG engineering guidelines
Network delay and packet loss evaluation example
From PING data, calculate the average one-waydelay (halved from PING
output and adding 93 ms IP Trunk3.01 (and later) processing delay) and
standard deviation for latency. Do a similar calculation for packet loss
without adjustment.
Adding a standard deviation to the mean of both delay and loss is for
planning purposes. A customer might want to know whether traffic
fluctuation in their intranet reduces the user’s QoS.
Table28 "Sample measurement results for G.729A codec" (page
146) provides a sample measurement of network delayand packet loss for
the G.729A codec between various nodes.
Tabl e 28
Sample measurement results for G.729A codec
Measured one-way
delay (ms)
Measured
Packet loss
(%)
Expected QoS level
(See Table 33 "IP Trunk
3.01 (and later) QoS
levels" (page 149))
Destination
pair Mean Mean+σMean Mean+σMean Mean+σ
Santa Clara/
Richardson 171 179 1.5 2.1 Excellent Good
Santa Clara/
Ottawa 120 132 1.3 1.6 Excellent Excellent
Santa Clara/
Tokyo 190 210 2.1 2.3 Good Good
Richardson/
Ottawa 220 235 2.4 2.7 Good Good
As an example, the delay and loss pair of traffic from Santa Clara to
Richardson (171 ms and 1.5%) will meet "excellent" criterion, but their
counter part with standard deviation (179 ms and 2.1%) can achieve only
"good" QoS.
Since the algorithm implemented in IP Trunk3.01 (and later) calculates only
mean and not standard deviation, it confirms the "excellent" rating (if the
objective is set forexcellent, it will not fallback to alternate facilities), but the
customer has up to a 50% chance of experiencing a service level inferior to
an "excellent" level.
Nortel Communication Server 1000
IP TrunkFundamentals
NN43001-563 02.01 Standard
Release 5.5 21 December 2007
Copyright© 2007, Nor tel Networks
.