Expand Configuration and Management Manual (G06.24+)
Tuning
Expand Configuration and Management Manual—523347-008
20-33
Example 2: Reducing Passthrough Traffic
number of processors, communications devices, and communications links used on
\JUICE as well as in the network.
Measuring Passthrough Traffic in an Entire Network
Example 20-8 shows another step in the complete analysis of passthrough traffic in an
Expand network. In this example, data taken from SCF is used to compute the total
network overhead of passthrough traffic for each source and destination at one node.
This single-node analysis could be extended to all nodes in a network.
Example 20-8 shows the true overhead produced by passthrough traffic for an entire
network. The route between \OAHU and \HERE consists of four hops because three
intermediate nodes and six Expand line-handler processes handle every message
between \OAHU and \HERE. For the network as a whole, 75 percent of the processor
overhead for traffic between \OAHU and \HERE is handled by the intermediate nodes.
Note. The table in Example 20-8 was prepared by combining information from Expand
subsystem SCF PATH STATS and PROBE commands and then manipulating the data with a
spreadsheet. The STATS command showed the traffic, while the PROBE command showed
the number of hops. Example 20-8
is an illustration of one simple technique for deriving a
thorough understanding of network performance from standard, readily available
instrumentation.
Example 20-8. Passthrough Traffic in a Network
Local System \HERE
Frames % Send % Send Rcv % Rcv
Remote Msgs Rcvd Frames Frames Total Pass Pass Pass Pass
System Sent Sent Rcvd Rcvd Hops Thru Thru Thru Thru
----------------------------------------------------------------------------
\OAHU 40856 87030 81732 9.87% 4 522180 10.14% 490392 14.37%
\CANTON 39259 79444 78637 9.50% 4 476664 9.25% 471822 13.83%
\SAIPAN 10438 76652 77156 9.32% 1 0 0.00% 0 0.00%
\WAKE 21888 90266 72997 8.82% 1 0 0.00% 0 0.00%
\MAUI 36327 73565 72830 8.80% 4 441390 8.57% 436980 12.81%
\HOME 33113 69507 66592 8.04% 4 417042 8.10% 399552 11.71%
\UIST 32536 67577 64460 7.79% 4 405462 7.87% 386760 11.33%
\ROSS 10401 28964 62856 7.59% 2 57928 1.12% 125712 3.68%
\ATTU 24123 113862 48640 5.87% 5 910896 17.68% 389120 11.40%
\ARAN 15023 42482 30128 3.64% 5 339856 6.60% 241024 7.06%
\JERSEY 3 55901 25063 3.03% 1 0 0.00% 0 0.00%
\BLOCK 4222 25134 23663 2.86% 1 0 0.00% 0 0.00%
\ALCA 269 56364 22189 2.68% 1 0 0.00% 0 0.00%
\TRAZ 3 51155 20649 2.49% 1 0 0.00% 0 0.00%
\KISKA 8236 71395 16628 2.01% 3 285580 5.54% 66512 1.95%
\HONSHU 7364 38922 15157 1.83% 4 233532 4.53% 90942 2.67%
\NOMAN 6938 14883 13932 1.68% 4 89298 1.73% 83592 2.45%
\MOGMOG 4603 27432 10078 1.22% 4 164592 3.19% 60468 1.77%
\CONEY 3229 32742 7161 0.86% 4 196452 3.81% 42966 1.26%
\KARGH 2706 26766 6313 0.76% 4 160596 3.12% 37878 1.11%
\ARMAGH 2617 27866 5568 0.67% 5 222928 4.33% 44544 1.31%
\GEDDON 2660 28414 5504 0.66% 5 227312 4.41% 44032 1.29%
========================================================================
Total 306814 1186323 827933 100% 71 5151708 100% 3412296 100%