User's Manual

240 APCD-LM043-4.0
• Violation counters could roll over after 70 seconds if the corresponding parameter was
set too small. A steeply climbing violation counter indicates serious problems with
either the settings or the system load.
To view the load statistics:
Console> radio meter
CCU Load Meter Time (us): 2388125184
Fw Pyls Fw Bytes Rev Pyls Rev Bytes Max Vio Ideal Vio
Gold Active 261996 117413346 233592 45583692 1 0
Gold Inactive 65145 3614618 3038 356335 0 0
Silver Active 885135 299817354 795036 131774471 0 2
Silver Inactive 50164 2246125 1283 172947 0 0
Bronze Active 20847 13538244 17727 2242378 0 0
Bronze Inactive 4098 178357 191 25156 0 0
BE Active 34294 2380443 3395 356844 0 97
BE Inactive 21443 708668 15808 688558 0 0
Broadcast 51436 2697875 48070 6848739 0 0
Overall 1394558 442595030 1118140 188049120 1 99
#EUMs Avg IPS Total Polls Empty Polls
Gold Active 0 21507 3017826 2577645
Gold Inactive 1 45867 2351419 2283292
Silver Active 0 1268 30527206 28936895
Silver Inactive 0 49800 1549140 1497712
Bronze Active 0 11906 257261 224260
Bronze Inactive 1 48111 125211 120926
BE Active 0 29463 707972 670633
BE Inactive 1 50472 1100549 1063487
Broadcast 1 1197 376559248 376459966
Console>
Each row on the above table, except for Overall, corresponds to a grade of service/activity
level combination. The last row is the overall totals, which in all cases except IPS, are the sum
of the entries in the column. Active refers to EUMs that have had traffic within the last
activePollTimeout, and Inactive refers to EUMs that have not had traffic within the last
activePollTimeout and have not timed out on disassociationTime. Therefore payload counts
for inactive EUMs indicate transitions from inactive to active.
NOTE: Broadcast traffic is carried with random access polls. Therefore,
forward traffic (CCU-to-EUM) in the broadcast row is broadcast
traffic, while reverse traffic is random access traffic. The reverse
packet rate should be less than 15% of the total poll rate for the
broadcast traffic, since random access attempts can collide.
Assuming uncorrelated arrivals, this should ensure that 99.7% of
random access attempts succeed (in four retries or less).