Cisco Cisco ASR 5000 Guía Para Resolver Problemas

In order to protect MME (imsimgr and sessmgr onwards), the throttling rate, queue wait time, and queue size
time should be defined. The throttling rate is dependent on the MME call model as the MME capacity is
dependent on the call model.

For MME the throttling rate calculation is relatively simple, take the standard Call Events Per Second (CEPS)
in the network plus tolerance.  Also, you might need to consider the HSS database capacity also if HSS
protection is needed.

In busy hours, MME handles up to 170 to 200 calls per second (Attaches+ Inter TAU). In case of one site
failure, up to 350 to 370 calls per second might arrive to one MME. Under this call rate, the MME utilization
rises close to 80% and 400 calls per second is an optimal level to limit the throttling rate in order to avoid
excessive signaling load inside the MME box.

The queue wait time by default is 5 seconds. It is optimal for CUSTOMER. The queue size by default is 2500.
It is optimal for CUSTOMER.

The configuration command is as follows.

asr5k(config)#network−overload−protection mme−new−connections−per−second

 new_connections action attach { drop | reject−with−emm−cause

 { congestion | network−failure | no−suitable−cell−in−tracking−area}

 tau { drop | reject−with−emm−cause { congestion | network−failure

 | no−suitable−cells−in−tracking−area | no−sec−ctxt−in−nw} fwd−reloc

 { drop | reject} }{wait−time  <wait−time>} {queue−size <queue−size>}

new_connections
Defines the number of new MME connections to be accepted per second. Must be an integer from 50 to 5000.
The default is 500.

action
Defines the action to be taken when the pacing queue becomes full. Whenever new connections are received
at the MME, they are queued in the pacing queue and imsimgr processes messages from the queue at the