There are 4 main sources of packet delay:

• $d_{proc}$ - nodal processing
  • this includes time for checking bit errors, determining output link, etc.
  • typically < microseconds
• $d_{queue}$ - queueing delay
  • this includes time waiting at the output link for transmission
  • depends on congestion level of router
• $d_{trans}$ - transmission delay
  • Essentially the time to push a packet with current bandwidth
  • $d_{trans} = \frac L R$, where $L$ is the packet length in bits, $R$ is the link transmission rate (bits/second)
• $d_{prop}$ - propagation delay
  • $d_{prop} = \frac d s$, where $d$ is length of physical link, and $s$ is propagation speed ($\approx 2 \times 10^8 m/s$)
  • can essentially be thought of the time for packet to travel (time to propagate)

Transmission delay and propagation delay are very different.

Note that with TDM/FDM, calculation is similar, but divide bandwidth by splits.

• Transmission time would be independent of number of links

Revisiting packet queueing delay, we can define “traffic intensity” as follows:

• $a$ - average packet arrival rate
• $L$ - packet length (bits)
• $R$ - link bandwidth (bit transmission rate)

$$ \frac{L \cdot a}{R} = \frac{\text{arrival rate of bits}}{\text{service rate of bits}} $$

• $La/R \approx 0$ ⇒ average queueing delay is small
• $La/R \approx 1$⇒ average queueing delay is large
• $La/R > 1$ ⇒ more packets (”work”) is arriving than can be serviced, so average delay is infinite

Traceroute