3com 3031 Instruccion De Instalación
978
C
HAPTER
70: L
INK
E
FFICIENCY
M
ECHANISMS
scheduled and waiting for being transmitted, it has to wait until all the large
packets have been transmitted. As for the real-time applications, large packets can
cause block and delay, consequently, the remote end cannot hear continuous
speech. It is required by the interactive voice that the end-to-end delay cannot be
larger than 100-150ms.
packets have been transmitted. As for the real-time applications, large packets can
cause block and delay, consequently, the remote end cannot hear continuous
speech. It is required by the interactive voice that the end-to-end delay cannot be
larger than 100-150ms.
Dispatching a large packet of 1500 bytes through a 56-kbps line, perhaps will take
215 ms, this will exceed the delay point that one can tolerate. LFI is a method for
fragmenting larger packets and adding both the smaller packets and fragments of
the large packet to the queue. The fragmented datagrams are reassembled at the
destination. LFI can reduce delay of real-time packets on relatively slow bandwidth
links.
215 ms, this will exceed the delay point that one can tolerate. LFI is a method for
fragmenting larger packets and adding both the smaller packets and fragments of
the large packet to the queue. The fragmented datagrams are reassembled at the
destination. LFI can reduce delay of real-time packets on relatively slow bandwidth
links.
The following figure describes the process of link fragmentation and interleaving.
When large packets and small voice packets arrives at an interface at the same
time, the large packets are fragmented into small fragments. If the interface is
configured with WFQ, the voice packets and these small fragments are interleaved
together and put into the WFQ.
When large packets and small voice packets arrives at an interface at the same
time, the large packets are fragmented into small fragments. If the interface is
configured with WFQ, the voice packets and these small fragments are interleaved
together and put into the WFQ.
Figure 242 Link fragmentation and interleaving
IP Header
Compression
Configuration
Compression
Configuration
IP Header Compression means to compress packet headers to reduce the total
length of the packet so as to increase the data transmission rate on the network. It
is mainly applied on PPP, HDLC and Frame Relay at the link layer. It's recommended
to apply on low speed link.
length of the packet so as to increase the data transmission rate on the network. It
is mainly applied on PPP, HDLC and Frame Relay at the link layer. It's recommended
to apply on low speed link.
IP Header Compression configuration includes:
■
Enable/disable IP Header Compression
■
Clear the cache entries of IP Header Compression (such as the context cache)
■
Configure the connection parameter of TCP compression
■
Configure the connection parameter of RTP compression
■
Monitor the information of TCP header compression
■
Monitor the information of IP Header Compression
Enabling/Disabling IP
Header Compression
Header Compression
Executing the command in the following table can enable the IP header
compression on some interface. If IP header compression is enabled, the TCP
compression on some interface. If IP header compression is enabled, the TCP
Large packet
Traffice
classifying
Sending queue
WFQ
Voice packet
Fragmentation