[netperf-talk] Interim results output enhancement for omni tests

[Rick Jones]

Folks -

I have enhanced the Interim results output for an "omni" or migrated
test (./configure --enable-demo --enable-omni and -D <mumble> as a
global command-line option) to include a "since the epoch" (gettimeofday
or what emulates it) timestamp to millseconds.  So what was once:

$ src/netperf -H 192.168.1.3 -l 12 -D 1 -t UDP_STREAM -- -m 1472 -s 64K
-S 64K
MIGRATED UDP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to
192.168.1.3 (192.168.1.3) port 0 AF_INET : histogram : demo
Interim result:  959.16 10^6bits/s over 1.04 seconds
Interim result:  956.91 10^6bits/s over 1.00 seconds
Interim result:  957.19 10^6bits/s over 1.00 seconds
Interim result:  956.98 10^6bits/s over 1.00 seconds
Interim result:  957.25 10^6bits/s over 1.00 seconds
Interim result:  957.24 10^6bits/s over 1.00 seconds
Interim result:  956.90 10^6bits/s over 1.00 seconds
Interim result:  956.92 10^6bits/s over 1.00 seconds
Interim result:  957.22 10^6bits/s over 1.00 seconds
Interim result:  956.91 10^6bits/s over 1.00 seconds
Interim result:  957.17 10^6bits/s over 1.00 seconds
Socket  Message  Elapsed      Messages                
Size    Size     Time         Okay Errors   Throughput
bytes   bytes    secs            #      #   10^6bits/sec

131072    1472   12.00      975475      0     957.24
131072           12.00      975475            957.24

Will become (sorry about the wrap)

$ src/netperf -H 192.168.1.3 -l 12 -D 1 -t UDP_STREAM -- -m 1472 -s 64K
-S 64K
MIGRATED UDP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to
192.168.1.3 (192.168.1.3) port 0 AF_INET : histogram : demo
Interim result:  959.16 10^6bits/s over 1.04 seconds ending at
1308181750.779
Interim result:  956.91 10^6bits/s over 1.00 seconds ending at
1308181751.781
Interim result:  957.19 10^6bits/s over 1.00 seconds ending at
1308181752.782
Interim result:  956.98 10^6bits/s over 1.00 seconds ending at
1308181753.782
Interim result:  957.25 10^6bits/s over 1.00 seconds ending at
1308181754.782
Interim result:  957.24 10^6bits/s over 1.00 seconds ending at
1308181755.782
Interim result:  956.90 10^6bits/s over 1.00 seconds ending at
1308181756.783
Interim result:  956.92 10^6bits/s over 1.00 seconds ending at
1308181757.783
Interim result:  957.22 10^6bits/s over 1.00 seconds ending at
1308181758.783
Interim result:  956.91 10^6bits/s over 1.00 seconds ending at
1308181759.783
Interim result:  957.17 10^6bits/s over 1.00 seconds ending at
1308181760.783
Socket  Message  Elapsed      Messages                
Size    Size     Time         Okay Errors   Throughput
bytes   bytes    secs            #      #   10^6bits/sec

131072    1472   12.00      975475      0     957.24
131072           12.00      975475            957.24


Which might make it easier to shove said interim results into something
like a round robin database (RRD) "after the fact" and have it line-up
reasonably well with timestamps of other RRDs on the system.

The output of the "classic" (vs migrated) tests remains unchanged.

happy benchmarking,

rick jones