Overload of packet core network
nodes in the network results in service de-gradation. Overload conditions can
occur in various network scenarios. Overload issue can be addressed through
improved load distribution over the network.
GTP-C load and overload control feature adds MME support for GTP-C
load and overload control mechanism on S11 interface. GTP-C load and overload
control is a standard driven (3GPP TS 29.807 V12.0.0 and 3GPP TS 29.274 V d30)
feature. For standards compliance information see the Standards Compliance
section in this feature chapter.
This feature is license controlled. The "EPC Support for GTP
Overload Control" license is required for successfully configuring and enabling
this feature. Please consult your Cisco Account Representative for information
about the specific license.
GTP-C Overload Issues and Resultant effects
A GTP-C overload occurs when the number of incoming requests exceeds
the maximum request throughput supported by the receiving GTP-C entity. The
GTP-C is carried over UDP transport, and it relies on the re-transmissions of
unacknowledged requests. When a GTP-C entity experiences overload (or severe
overload) the number of unacknowledged GTP-C messages increase exponentially
and this leads to a node congestion or even node collapse. An overload or
failure of a node further leads to an increase of the load on the other nodes
in the network and in some cases into a network issue.
Listed below are some examples of GTP-C signaling based scenarios
which lead to GTP-C overload:
A traffic flood resulting from the failure of a network element,
inducing a signaling spike.
A traffic flood resulting from a large number of users performing
TAU/RAU or from frequent transitions between idle and connected mode.
An exceptional event locally generating a traffic spike for
example a large amount of calls (and dedicated bearers) being setup almost
Frequent RAT re-selection due to scattered Non-3GPP (for example,
Wi-Fi) coverage or a massive mobility between a 3GPP and Non-3GPP coverage.
This may potentially cause frequent or massive inter-system change activities.
GTP-C overload may result in any of the following service impacts:
Loss of PDN connectivity (IMS, Internet and so on) and associated
Loss of ability to setup and release radio and core network
bearers necessary to support services, for example GBR bearers for VoLTE.
Loss of ability to report to the PGW/PCRF user's information
changes, for example location information for emergency services and lawful
intercept, changes in RAT or QoS.
GTP-C load control
and overload control are complimentary concepts which can be supported and
activated independently on the network. This feature uses the existing EGTPC
infrastructure to gather and distribute load and overload control information
across session managers. In broad terms GTP-C load control can be described as
a preventive action and GTP-C overload control can be described as a corrective
action. A GTP-C entity is termed as overloaded when it operates over and above
its signaling capacity resulting in a diminished performance (including impacts
to handling of incoming and outgoing traffic).
of enabling GTP-C load control are listed below:
allows better balancing of the session load; this prevents an GTP-C overload
enables a GTP-C entity (for example SGW or PGW) to send its load information to
a GTP-C peer (for example a MME or SGSN, ePDG, TWAN) to adaptively balance the
session load across entities supporting the same function (for example SGW
cluster) according to their effective load. The load information reflects the
operating status of the resources of the GTP-C entity.
does not trigger overload mitigation actions even if the GTP-C entity reports a
of enabling GTP-C overload control are listed below:
prevents a GTP-C entity from becoming or being overloaded to gracefully reduce
its incoming signaling load by instructing its GTP-C peers to reduce sending
traffic according to its available signaling capacity to successfully process
aims at shedding the incoming traffic as close to the traffic source as
possible when an overload has occurred.
Benefits of GTP-C Load and Overload Control Support on MME:
distribution on SGW and PGW this in turn reduces the occurrence of PGW/SGW
pro-actively advertises its overload information so that the peer nodes SGW/PGW
can reduce the traffic.
The MME can
reduce the traffic towards the peer SGW/PGW if they are overloaded.
This feature utilizes the existing
EGTPC infrastructure to gather and distribute load and overload control
information across session managers.
How it Works
This section describes the detailed
working mechanism of this feature.
Load or Overload Control Profile
associate command associates or disassociates
supportive services and policies with an MME service. New keywords
gtpc-overload-control-profile are introduced in the
associate command to configure the GTP-C load control
profile and GTP-C overload control profile.
gtpc-load-control-profile is used to configure GTP-C
Load Control Profile for this MME service.
gtpc-overload-control-profile is used to configure
GTP-C Overload Control Profile for this MME service.
profile_name is a string of size 1 up to 64.
commands are not enabled by default.
Configuring Usage of
GTP-C Load Information in SGW/PGW Selection
command configures the parameters controlling the gateway selection process. A
gtp-weight is introduced as a part of this feature
which is the weight value calculated from the Load Control Information received
from the GTP peers.
enables the MME selection of SGW and PGW based on the advertised load control
configuration can be applied selectively to subscribers.
This CLI is not
enabled by default.
Manager and IMSI Manager CPU Utilization to Calculate Overload Factor
This new command allows the user to
configure the inclusion of CPU utilization of Session Manager, Demux Manager,
IMSI Manager and MME Manager under GTP-C overload control profile for overload
show configuration command to verify the
configuration, the part of the output related to GTP-C overload control profile
Configuration displays the following parameters based on the configuration:
cpu-utilization is used to configure the inclusion
of CPU utilization of Session Manager, Demux Manager, IMSI Manager and MME
Manager under GTP-C overload control profile for overload factor calculation.
Troubleshooting the GTP-C Load and Overload Control Feature
This section provides information
regarding show commands and bulk statistics for this feature.
configuration to consider LCI/OCI information in node selection is enabled. For
more information see the section “Configuration to use GTPC load information in
SGW/PGW selection” in this feature chapter. Execute the show command
call-control-profile full all to verify the same.
For the GTP
dynamic weight (that is, LCI/OCI) to work in case of DNS based node selection,
following DNS weight based configuration should be present in the call control
If the MME is not
reporting overload control information, follow the steps described below:
Step 1 is working as explained earlier in this section.
the overload control profile is associated with the MME Service. For more
information see the section “Configuring GTP-C Load or Overload Control
Profile” in this feature chapter.
following show commands to verify if the feature is enabled:
show mme-service all
show egtp-service name egtp_mme
parameters configured through the commands inclusion-frequency, message-
prioritization, overload-control-handling, overload-control-publishing,
self-protection-behavior, tolerance, throttling-behavior, validity-period and
weightage under the GTPC Overload Profile Configuration Mode. Execute the
following show command to verify the same: