Information Technology Reference
In-Depth Information
Case Study Summary
This case study examined the input drops in the particular test scenarios where input drops
occurred. The number of input drops provided insight into the number of BGP packets
being generated. The larger the number of BGP updates, the more incoming TCP ACKs
there are. Figure 3-14 shows the input hold queue drops.
Figure 3-14 Final Summary Input Drop Counts
Final Summary - Input Drop Counts
1600
1400
1200
1000
800
600
400
200
0
70, 000
80,000
90, 000
100,000
110, 000
120,000
130, 000
140,000
Base Line
202
264
257
264
252
464
366
467
Peer G roups
705
775
854
904
1076
1144
1263
1343
Peer G roups, PMTUD
122
93
116
134
139
190
213
431
Prefix Count
The major point from the input drop counts is the massive increase when peer groups are
enabled. This increase indicates the impact that peer groups have on increasing update
generation through replication.
Figure 3-15 shows the convergence times for all six test scenarios to summarize the impact
of BGP convergence tuning and the interdependencies between the various aspects of
convergence tuning.
Throughout this case study, the improvements in convergence time have been discussed as
percentages. Table 3-3 compares all the prefix counts and the relevant scenarios.
 
Search WWH ::




Custom Search