100% found this document useful (1 vote)
312 views

Load MNG PDF

Uploaded by

Katea Ulianovici
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
100% found this document useful (1 vote)
312 views

Load MNG PDF

Uploaded by

Katea Ulianovici
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 33

Basic Load Management

Feature Description

54/221 04-LZA 701 6014/1 Uen A1


Copyright

© Ericsson AB 2016. All rights reserved. No part of this document may be


reproduced in any form without the written permission of the copyright owner.

Disclaimer

The contents of this document are subject to revision without notice due to
continued progress in methodology, design and manufacturing. Ericsson shall
have no liability for any error or damage of any kind resulting from the use of
this document.

54/221 04-LZA 701 6014/1 Uen A1 | 2016-01-05


Contents

Contents

1 Introduction 1
1.1 Basic Characteristics 1
1.2 Basic Load Management Summary 1

2 Feature Operation 3
2.1 Network Requirements 3
2.2 Paging Intensity Control 3
2.3 DU Connection Intensity Control 4
2.4 Static Cell Connection Intensity Control 6
2.5 MP Load Control 7
2.6 Procedure Latency Supervision 14
2.7 Access Class Barring 15
2.8 BB Management Interface Intensity Control 16

3 Parameters 18
3.1 Feature Configuration Parameters 18
3.2 Affected Parameters 19

4 Network Impact 20
4.1 Capacity 20

5 Associated Features and Affected Functions 21


5.1 Prerequisite Features 21
5.2 Affected Features 21
5.3 Related Features 21
5.4 Affected System Functions 21

6 Performance 22
6.1 KPIs 22
6.2 Counters 23
6.3 Events 27

7 O&M Information 29
7.1 Feature Activation 29

54/221 04-LZA 701 6014/1 Uen A1 | 2016-01-05


Basic Load Management

54/221 04-LZA 701 6014/1 Uen A1 | 2016-01-05


Introduction

1 Introduction

This document describes the Load Management factors that impact eNodeB
capacity and the functions used to manage it. This includes the functioning
and interactions of the load control and overload protection mechanisms and
their impact on Key Performance Indicators (KPI).

The purpose of Basic Load Management is to protect the eNodeB and it's cells
from overload while minimizing the impact to the traffic.

1.1 Basic Characteristics


This section describes the basic characteristics of the feature.

Feature name: Basic Load Management.

Product identity: Radio Node LTE FAJ 121 3092

Replaces: N/A

Dependencies

This feature has no prerequisite features and affects no other features.

1.2 Basic Load Management Summary


The benefits of Basic Load Management are as follows:

• Reduced risk for node crash.

• Efficient resource allocation within the Main Processor (MP) to ensure that
traffic and Operation and Maintenance (O&M) procedures are not delayed
too long.

• Maximized traffic throughput at loads above engineered capacity.

• Improved performance allowing increased MP load before traffic rejection.

• Avoiding cells being stuck in an overloaded state due to air interface


congestion.

• Improved Retainability through overload protection.

• Improved accessibility for high priority calls (for example Emergency, High
Priority Access) during high MP load.

54/221 04-LZA 701 6014/1 Uen A1 | 2016-01-05 1


Basic Load Management

• Overload protection against signalling storms.

2 54/221 04-LZA 701 6014/1 Uen A1 | 2016-01-05


Feature Operation

2 Feature Operation

This section describes the Basic Load Management feature in more detail,
including network configuration requirements, operation flows and factors that
impact MP load.

The feature consists of the following self-contained functions, described


separately throughout the document:

• Paging Intensity Control

• DU Connection Intensity Control

• Static Cell Connection Intensity Control

• MP Load Control

• Procedure Latency Supervision

• Access Class Barring

• BB Management Interface Intensity Control

Privileged access, such as Emergency and High Priority Access (HPA), are
allowed to connect at all times, except when MP Load Control is in
OVERLOAD state.

2.1 Network Requirements


This feature has no prerequisite network configuration requirements.

2.2 Paging Intensity Control


To guarantee that the paging function in eNodeB always works as intended, a
load protection mechanism is included, that limits the number of paging
messages processed within a certain time. The load protection mechanism
protects both the MP as well as BB resources on eNodeB and Cell levels.

During normal operation the Paging Intensity Control ensures that only a
predefined number of paging messages will be handled within a certain time
period.

Figure 1 and Figure 2 provide overviews of the Paging Intensity Control.

54/221 04-LZA 701 6014/1 Uen A1 | 2016-01-05 3


Basic Load Management

Measurements Paging

eNB Paging Intensity Cell Paging Intensity


Window Window
Triggers

Actions Discard Accept Discard Accept

L0000940A

Figure 1 Paging Intensity Control (protecting MP and BB)

Allow Q paging requests during TP

x x xx x x x x xxxxxx x x x x x x xx
time

Legend: ∑x ≤ Q : request allowed


x = Paging request ∑x > Q : request rejected

L0000941A

Figure 2 Paging Intensity Window


As long as the paging intensity in either the eNodeB or the cell is below or
equal to the maximum number of allowed paging messages for the current
period, the operation is accepted and sent to the UE. Otherwise the paging is
discarded.

The MP Load Control mechanism can adjust the number of allowed paging
requests.

2.3 DU Connection Intensity Control


To guarantee that the connection handling in eNodeB always works as
intended, a load protection mechanism is included, that limits the number of
simultaneous initial establishments and incoming handovers processed within
a certain time. The load protection mechanism protects both the MP as well as
BB resources on DU level.

4 54/221 04-LZA 701 6014/1 Uen A1 | 2016-01-05


Feature Operation

During normal operation the DU Connection Intensity Control ensures that


only a predefined number of initial establishments and incoming handovers
will be handled within a certain time period.

Figure 3 and Figure 4 provide overviews of the DU Connection Intensity


Control.

Measurements Initial Establishment Incoming Handover

DU Connection
Intensity Windows
Triggers

Actions Accept Reject

L0000942A

Figure 3 DU Connection Intensity Control

Allow Z connection requests during TL Long Window


(protecting MP)

x xx x x x x x x xxxxx xxxxx xx x x
x xx x x x x x x xxxxx x xxxxx xx
time

Allow Y request messages during TS Short Window


(protecting BB)

Legend: ∑x ≤ Z | Y : request allowed


x = Connection request ∑x > Z | Y : request rejected
L0000943A

Figure 4 DU Connection Intensity Window

54/221 04-LZA 701 6014/1 Uen A1 | 2016-01-05 5


Basic Load Management

As long as the initial establishment and incoming handover intensity in the DU


is below or equal to the maximum number of allowed connections for the
current period, the connection is accepted and set up. Otherwise the
connection is rejected.

The MP Load Control mechanism can adjust the number of allowed initial
establishments and incoming handovers by the short and long DU Connection
Intensity Windows, DCIWS and DCIWL.

Note: Privileged calls, such as High Priority Access (HPA) and Emergency,
are accepted at all times, regardless of DU connection intensity.

2.4 Static Cell Connection Intensity Control


To guarantee the radio resource utilization in the cell, a load protection
mechanism is included. The load protection mechanism ensures that the
already connected user's quality of service is maintained on cell level by
allowing only a predefined number of initial establishments and incoming
handovers to be handled within a certain time period.

Figure 5 and Figure 6 provide overviews of the Static Cell Connection


Intensity Control.

Measurements Initial Establishment Incoming Handover

Cell Connection
Intensity Window
Triggers

Actions Accept Reject

L0000936A

Figure 5 Static Cell Connection Intensity Control

6 54/221 04-LZA 701 6014/1 Uen A1 | 2016-01-05


Feature Operation

Allow N connection requests during Tc

x x xx x x x x xxxxxx x x x x x x xx time

Legend: ∑x ≤ N : request allowed


x = Connection request ∑x > N : request rejected

L0000937A

Figure 6 Static Cell Connection Intensity Window


As long as the initial establishment and incoming handover intensity in the cell
is below or equal to the maximum number of allowed connections for the
current period, the connection is accepted and set up. Otherwise the
connection is rejected.

The cell connection intensity control mechanism can adjust the number of
allowed initial establishments and incoming handovers by the Cell Connection
Intensity Window, CCIW.

Note: Privileged calls, such as High Priority Access (HPA) and Emergency,
are accepted at all times, regardless of cell connection intensity.

2.5 MP Load Control


This section describes the MP Load Control function in more detail.

2.5.1 Basic Principles


MP Load Control strives to maintain the load below a target value and regulate
the processor load on a per Digital Unit (DU) level.

In order for the eNodeB to utilize its resources in an optimal way and avoid
overload, MP Load Control is designed using the following main principles:

1. Allocation of resources is performed as efficiently as possible.

2. The Central Processing Unit (CPU) overhead contributed by the operation


of the MP Load Control mechanism is kept to a minimum.

3. The system protects itself from overload by rejecting activities when


processing resources are busy.
MP Load Control uses the load control and overload protection
mechanisms to fulfill these overall goals.

54/221 04-LZA 701 6014/1 Uen A1 | 2016-01-05 7


Basic Load Management

Note: MP Load Control regulate the incoming traffic intensity and is not
related to the amount of UE that are connected to the eNodeB.
MP Load Control can therefore reject incoming traffic even if the
licensed number of users is not reached.

2.5.2 Load Control and Overload Protection


The load control mechanism works within an MP in a single-core or multi-core
environment as well as in single DU or multi DU configurations. Traffic
management software negotiates with load control during the initial phases of
establishments, incoming handovers and incoming paging. In the Digital Unit
multiStandard (DUS), the MP cores are protected by the load control
mechanism by using measurements from the highest loaded core.

The MP Load Control algorithm strives to keep the processor load from
exceeding the target load level. New Radio Resource Control (RRC)
connection setups, incoming handovers and pagings request the required
capacity for load control before proceeding. The load management algorithm
dynamically sets the permitted intensity limits for RRC connection setups,
incoming handovers and pagings based on continuous CPU load
measurements. When CPU load measurements indicate that the CPU load is
higher than the target value, the permitted intensity limits are decreased.
When CPU load measurements indicate that the CPU load is lower than the
target value, the permitted intensity limits are increased.

Figure 7 provides an overview of how the load control mechanism regulates


certain activities in different states using a state variable.

8 54/221 04-LZA 701 6014/1 Uen A1 | 2016-01-05


Feature Operation

MP Processor
Measurements Load

MP Load State

Not high
Triggers
High Very high Overload

Reject RRC
Actions Modify Paging Modify DU Connection
Connection Request and
Intensity Window Intensity Window
Incoming Handover

L0000944A

Figure 7 MP Load Control Mechanism


Measurements of the CPU load are fetched every tenths of a second through
the capacity measurement interface. The average CPU load is calculated over
one second within MP Load Control. In the DUS, the CPU load from the
highest loaded core is measured.

Four MP Load Control states can be derived from the MP CPU load. See
Table 1 .

Table 1 MP Load Control States


MP CPU Load MP Load Control State
Greater than 98% OVERLOAD
90% to 98% VERY HIGH
70% to 89% HIGH
Less than 70% NOT HIGH

Overload protection actions and their severity depends on the MP Load


Control state.

These actions are designed to adhere to the following main principles:

• Reject all activities in state OVERLOAD.

54/221 04-LZA 701 6014/1 Uen A1 | 2016-01-05 9


Basic Load Management

• Reject part of the traffic in state VERY HIGH.

• Reject activities that do not directly impact traffic in state HIGH.

• No load control activities are required in state NOT HIGH.

Note: Due to for example bursts in incoming traffic, rejects can also happen
in state HIGH and NOT HIGH if the incoming traffic intensity exceeds
the permitted intensity limit.

The required action taken for each MP Load Control state is described in
Table 2 .

Table 2 MP Load Control Actions


MP Load Control Paging RRC Connection
State Setups and
Handovers
OVERLOAD Discards all incoming Rejects all RRC
paging. connection setups and
incoming handovers.
VERY HIGH Rejects more pagings than Rejects more RRC
in the last measurement connection setups and
interval by decreasing the more incoming
permitted intensity limit. handovers than in the
previous measurement
interval by decreasing
the permitted intensity
limit.
HIGH If the permitted intensity limit If the permitted
is not at its maximum size, it intensity limit is not at
is increased at each its maximum size, it is
measurement interval until it increased each
reaches the maximum size. measurement interval
until it reaches the
maximum size.
NOT HIGH Performs overload Performs overload
protection using the protection using the
maximum size of the maximum size of the
permitted intensity limit. permitted intensity
limit.

Note: RRC Connection Requests corresponding to Emergency and High


Priority Access (HPA) are allowed regardless of the permitted intensity
when the MP Load is in NOT HIGH, HIGH, or VERY HIGH state.

In state OVERLOAD all RRC Connection requests are rejected


including Emergency and HPA.

10 54/221 04-LZA 701 6014/1 Uen A1 | 2016-01-05


Feature Operation

Load control and overload protection of incoming traffic is done to protect the
eNodeB from being overloaded. In normal operation, the permitted intensity
limits are dynamically adjusted every 800 ms based on the current MP Load
Control state. The permitted intensity limit is checked before each load-
controlled activity is permitted to be executed. If the limit has been reached,
the activity is not permitted to start execution and is rejected. Furthermore, the
load control mechanism is designed to allow CPU time for lower priority and
background activities in order to avoid starvation of O&M activities.

2.5.3 Optimizing MP Load


This section shows the factors that impact the MP load and suggestions to
optimize them.

The largest contributors to MP load are service requests, releases, handovers,


pagings and event handling. Therefore, the focus is on optimizing factors that
impact these UE handling functions. Some of the key factors to optimize are
summarized below.

54/221 04-LZA 701 6014/1 Uen A1 | 2016-01-05 11


Basic Load Management

Traffic Behavior An increase in signaling increases the MP load. It is


recommended to reduce signaling activity by optimizing
the air interface. A radio network with good conditions
reduces the signaling activity compared to a radio
network with impaired radio conditions. A higher
inactivity timer value also decreases the signaling load
on the air interface, MP and eNodeB.

User Equipment UE behavior impacts the MP load. UE applications that


(UE) poll servers such as email applications, increase
signaling which in turn increases MP load. Typically,
smartphones use such applications more frequently
than other devices, which results in an increased MP
load. Optimizing the radio conditions for the UE and
traffic types helps reduce the MP load.

Paging Paging is used by the network to communicate with the


UE when in idle mode. Typically, a page is broadcast in
all cells belonging to the Tracking Area (TA) in which
the UE resides. Reducing the size of the TA reduces
the number of cells being paged, and thereby reduces
the CPU load in the MP.

Performance Collection of PM counters and event logs such as UE


Management (PM) traces increases MP load. By turning off non-essential
Counters and PM mechanisms, it is possible to reduce MP load.
Events

Digital Unit (DU) The MP capacity on the Digital Unit multiStandard


(DUS) is higher than the MP capacity on the Digital Unit
LTE (DUL). Therefore, to increase MP capacity, DUL
should be replaced by DUS where possible.

Handover Cells with a high number of handovers increase the MP


load. Reducing the number of handover requests
through optimization reduces the MP load.

Radio Information System information of other Radio Access


Methods (RIM) Technologies (RATs) are kept updated using RIM.
Using incorrectly configured RIM procedures increases
the MP load. The RIM procedures should always be
optimally configured.
To reduce the load on the MP in addition to these key factors, optimization of
RAN parameters, eNodeB functions and baseband (BB) functions should be
regularly performed.

2.5.4 Optimizing MP Load on Multi-Core Capable Digital Units


The first generation of DUs for macro eNodeBs, for example, the DU for LTE
(DUL), have a single-core MP processing unit, whereas later generations, for
example, DU for multiStandard (DUS) have a multi-core MP processing unit.

12 54/221 04-LZA 701 6014/1 Uen A1 | 2016-01-05


Feature Operation

In single-core DUs, all MP load modules are run in the available core, whereas
in multi-core DUs the MP load modules are distributed over the different cores,
to optimize performance as well as robustness.

Note: Certain load modules are statically allocated to a specific core,


whereas some load modules are dynamically allocated to one or more
cores that can change over time.

The largest contributors to MP load are UE related operations such as service


requests, releases, handovers, pagings, and event handling. These operations
are mainly realized by load modules cellLm, centralLm and monitorLm and
at high signaling intensities these load modules consume a major part of the
available CPU cycles.

• Loadmodule centralLm primarily handles signalling towards eNodeB


external nodes, for example S1 and X2, and interwork with loadmodule
cellLm.

• Loadmodules monitorLm and distmonLm primarily handle Performance


Management operations, that is counter and event generation.

• Loadmodule cellLm primarily handles UE signalling and is from an


execution point of view structured in a number of processes where each
process represents a set of cells. Each process is allocated to a
processing core and the cells are placed in a process is determined by
configuration data of expected cell traffic load, see Section 3.1.4 on page
18.

In a single DU eNodeB with multi-core support the cellLm, centralLm and


monitorLm load modules execute distributed over it's cores. In a DUS-41 the
cellLm load will be spread over two processes.

In a multi DU eNodeB with multi core support the cellLm, centralLm and
monitorLm execute on the master DU in the same manner as a single DU
eNodeB. The extension DU(s) run cellLm and distmonLm where the cellLm
load in a DUS-41 will be spread over three processes.

The master-DU hosts all external network traffic including management


operations in the eNodeB leaving the extension DU(s) to primarily handle cell
traffic load. Due to this the master DU cannot handle the same cell traffic load
as the extension DUs. As a rule of thumb, the master DU can handle
approximately two thirds of the cell traffic load that an extension DU can
handle.

In order to achieve maximum performance from the multi-core MP processing


capable eNodeB in either single or multi DU configurations, certain criteria
must be observed:

• The number of cells allocated per process is dependent on how the cell
traffic load weight is set for the cells in the eNodeB, see Section 3.1.4 on
page 18. With certain settings there can be an equal amount of cells per

54/221 04-LZA 701 6014/1 Uen A1 | 2016-01-05 13


Basic Load Management

process and with other settings it will be different number of cells per
process.

• The cell traffic load weight will not be able to act on traffic load distribution
over time. This implies that if two cells have the same maximum number
of expected RRC connected users, for example, the busy hours are in the
morning in one cell and in the evening in another cell, it is possible to
manually configure the traffic load weight so that the cells can be placed in
one process provided they run within one DU.

• The cell traffic load in the eNodeB must be distributed between the DUs in
a multi DU eNodeB configuration, so that a proper amount of traffic load is
placed on each DU. The master DU can not carry as much load as the
extension DU(s).

• The cell traffic load in the eNodeB must be distributed over the cells in a
multi-core capable DU so that a similar amount of traffic load is placed on
the processes of the cellLm load module.

This is primarily controlled by cell planning but can also be assisted by Inter
Frequency Load Balancing and Traffic Management features.

2.6 Procedure Latency Supervision


A load control mechanism used to protect each cell and its air interface is
Procedure Latency Supervision. A number of RRC procedures are time-
supervised and statistics of the measured time of these procedures are
collected. When this measurement indicates that the average latency of the
procedures is above a threshold, overload protection actions are taken. A high
latency of these RRC procedures indicate that the cell air interface is too busy
and rejection of new traffic is necessary in order to restore normal operation.
The Procedure Latency Supervision operates on a per Cell level.

Figure 8 provides an overview of Procedure Latency Supervision.

14 54/221 04-LZA 701 6014/1 Uen A1 | 2016-01-05


Feature Operation

Latency
Measurements Supervision

Procedure Latency
Supervision State
Triggers
Normal Blocking

Actions Reject RRC Connection Request


and Incoming Handover
L0000945A

Figure 8 Procedure Latency Supervision


The supervised RRC procedures are as follows.

• RRC Security Mode Setup

• RRC User Equipment (UE) Capability Enquiry

• RRC Connection Reconfiguration

• RRC Connection Setup

When, the procedure latency supervision state is “Blocking”, the action taken
is to reject all incoming traffic requests for a period of time. When that time has
expired, new incoming traffic requests are permitted again and the RRC
procedures are time supervised again.

2.7 Access Class Barring


To prevent bad accessibility, or bad retainability, or low throughput, or all, in an
overloaded cell, Access Class Barring can manually be configured to
momentarily limit the number of initial establishments in the cell. Access Class
Barring information is broadcasted in SIB2.

Figure 9 provides an overview of Access Class Barring.

54/221 04-LZA 701 6014/1 Uen A1 | 2016-01-05 15


Basic Load Management

Overloaded Cell

Observation Bad accessibility Low throughput

Bad retainability

Activate
Actions Access Class Barring
(Manual MOM configuration)
L0000947A

Figure 9 Access Class Barring


For more information about Access Class Barring, see Idle Mode Support

2.8 BB Management Interface Intensity Control


The BB Management Interface Intensity Control mechanism protects the
baseband control logic from overload.

In normal situations the other load control mechanisms will reject at high RRC
signalling intensities so the triggering of this load control mechanism is to
preserve a robust behavior in extreme situations. The BB Management
Interface Intensity Control operates on a per DU level.

Figure 10 provides an overview of BB Management Interface Intensity


Control.

16 54/221 04-LZA 701 6014/1 Uen A1 | 2016-01-05


Feature Operation

Measurements eNB internal signals originating from RRC procedures

eNB internal
signal sequence
Triggers

Actions Accept Reject


L0000948A

Figure 10 BB Management Interface Intensity Control

54/221 04-LZA 701 6014/1 Uen A1 | 2016-01-05 17


Basic Load Management

3 Parameters

This section describes parameters introduced by the Basic Load Management


feature and parameters affected by activating the feature.

Note: It is not possible for the operator to change the thresholds that control
Basic Load Management.

3.1 Feature Configuration Parameters


This section describes configuration parameters for the functions in Basic
Load Management.

3.1.1 Paging Intensity Control


Paging Intensity Control introduces no new parameters.

3.1.2 DU Connection Intensity Control


DU Connection Intensity Control introduces no new parameters.

3.1.3 Static Cell Connection Intensity Control


Static Cell Connection Intensity Control introduces no new parameters.

3.1.4 MP Load Control


Table 3 lists the parameters related to cell allocation on multi-core DUs.

Note: The parameters exist in the following MO classes:

• EUtranCellFDD

• EUtranCellTDD

Table 3 Parameters - MP Load Control


Parameter Description
expectedMaxNoOfUsers The maximum expected number of UE in
InCell RRC_CONNECTED mode in the cell. Will be used at unlock
of the cell to distribute cells to different cores in Digital Units
with a multi core Main Processor. A value of -1 indicates that
measuredMaxNoOfUsersInCell will be used to
control the cell distribution.

18 54/221 04-LZA 701 6014/1 Uen A1 | 2016-01-05


Parameters

Parameter Description
measuredMaxNoOfUsers Peak number of UE in RRC_CONNECTED mode measured
InCell in the cell since last restart. This attribute will be updated during
operation of the eNB and used at unlock of the cell to distribute
cells to different cores in Digital Units with a multi core Main
Processor. This attribute will not be used if MOM attribute
expectedMaxNoOfUsersInCell is set to zero or
higher.

More information about counters can be found in Managed Object Model


(MOM).

3.1.5 Procedure Latency Supervision


Procedure Latency Supervision introduces no new parameters.

3.1.6 Access Class Barring


For information about Access Class Barring parameters, see Idle Mode
Support

3.1.7 BB Management Interface Intensity Control


BB Management Interface Intensity Control introduces no new parameters.

3.2 Affected Parameters


The implementation of this feature affects no parameters.

54/221 04-LZA 701 6014/1 Uen A1 | 2016-01-05 19


Basic Load Management

4 Network Impact

This section describes how the Basic Load Management feature impacts the
network functions and capabilities.

4.1 Capacity
This section provides information on the impact of the feature on network
capacity.

4.1.1 MP Load Control


The MP capacity of the DUS is higher than the MP capacity of the DUL and
the DUS-41 capacity is higher than DUS-31.

20 54/221 04-LZA 701 6014/1 Uen A1 | 2016-01-05


Associated Features and Affected Functions

5 Associated Features and Affected


Functions

This section describes how the Basic Load Management feature affects other
features and functions.

5.1 Prerequisite Features


This feature has no prerequisite features.

5.2 Affected Features


This feature affects no other feature.

5.3 Related Features


Table 4 lists features that are related to the Basic Load Management feature.

Table 4 Related Features


Feature Description
Dynamic Load Control When Dynamic Load Control is activated Basic Load
Management: Procedure Latency Supervision is automatically
deactivated
Priority Paging When the core network is configured for prioritization of S1AP
paging messages, the Priority Paging feature must be enabled
to ensure that paging messages with a higher priority get a
higher success rate than paging messages with a lower
priority.
Idle Mode Support Manual operation of Access Class Barring is
described in Idle Mode Support.
Load-Based Access Barring The Load-Based Access Barring feature sets the barring
information broadcasted in the SIB2 based on the load control
levels in a cell. When the Load-Based Access Barring feature
is activated for a certain traffic type, manually configured
Access Class Barring as described in Idle
Mode Support is disabled for that traffic type.

5.4 Affected System Functions


When Basic Load Management mechanisms act in overload situations,
incoming handovers and initial establishment can be rejected.

54/221 04-LZA 701 6014/1 Uen A1 | 2016-01-05 21


Basic Load Management

6 Performance

This section describes performance indicators, counters, and events


associated with the Basic Load Management feature.

6.1 KPIs
This section describes the KPIs for the different functions in Basic Load
Management.

6.1.1 Paging Intensity Control


This function has no associated Key Performance Indicators (KPIs).

6.1.2 DU Connection Intensity Control


This function has no associated Key Performance Indicators (KPIs).

6.1.3 Static Cell Connection Intensity Control


This function has no associated Key Performance Indicators (KPIs).

6.1.4 MP Load Control


This function has no associated Key Performance Indicators (KPIs) however
MP Load Control prioritizes the retainability Key Performance Indicator
(KPI) over the accessibility KPI.

More information about KPIs can be found in Key Performance Indicators.

6.1.5 Procedure Latency Supervision


This function has no associated Key Performance Indicators (KPIs).

6.1.6 Access Class Barring


For information about Access Class Barring KPI's, see Idle Mode Support

22 54/221 04-LZA 701 6014/1 Uen A1 | 2016-01-05


Performance

6.1.7 BB Management Interface Intensity Control


This function has no associated Key Performance Indicators (KPIs).

6.2 Counters
This section describes the counters for the different functions in Basic Load
Management.

6.2.1 Paging Intensity Control


Table 5 lists the counters associated with Paging Intensity Control.

Table 5 Counters - Paging Intensity Control


Counter Description
pmPagS1Received This counter counts the number of received S1AP paging
messages in the RBS.
pmPagS1Discarded The number of discarded S1AP paging messages not routed to
any cell due to that the intensity exceeds the permitted intensity
limit.

More information about counters can be found in Managed Object Model


(MOM).

6.2.2 DU Connection Intensity Control


Table 7 and Table 8 lists the counters associated with DU Connection
Intensity Control.

The counters in Table 7 exist in the following MO classes:

• EUtranCellFDD

• EUtranCellTDD

Table 7 Counters - DU Connection Intensity Control


Counter Description
pmRrcConnEstabFailMp This counter is stepped for each rejected RRC Connection
Overload setup triggered by the MP Overload state.
pmRrcConnEstabFailDu This counter is stepped for each failed RRC Connection
Intens Establishment due to the DU Connection Intensity Control.
pmHoPrepRejInMpOverl This counter is stepped for each failed handover preparation
oad triggered by the MP overload state.
pmHoPrepRejInDuInten This counter is stepped for each failed Handover preparation
s due to DU Connection Intensity Control.

54/221 04-LZA 701 6014/1 Uen A1 | 2016-01-05 23


Basic Load Management

Counter Description
pmAdjustAccessMpLoad This counter shows how long time within the ROP period that
Ctrl the DU load control window for new accesses (RRC
Connection Establishment and Handover) has been down
regulated from its maximum value due to high MP load.

The counters in Table 8 exist in the MO class MpProcessingResource and are


also associated with DU Connection Intensity Contro.l

Table 8 Counters - DU Connection Intensity Control


Counter Description
pmAdjustAccessMpLoadCtrlDistr This counter is stepped periodically for the DU Connection
Intensity Control and shows the distribution of the allowed RRC
Connection Establishment and Handover for each time interval.
pmDuIntensHoRrcConnReqDistr This counter is stepped periodically and shows the distribution
of the Handover Request and RRC Connection Request
intensity in relation to the DU Connection Intensity Control limit.
pmDuIntensHoReqDistr This counter is stepped periodically and shows the distribution
of the Handover Request intensity in relation to the DU
Connection Intensity Control limit.
pmDuIntensRrcConnReqDistr This counter is stepped periodically and shows the distribution
of the RRC Connection Request intensity in relation to the DU
Connection Intensity Control limit.

More information about counters can be found in Managed Object Model


(MOM).

6.2.3 Static Cell Connection Intensity Control


Table 11 lists the counters associated with Static Cell Connection Intensity
Control.

Note: The counters exist in the following MO classes:

• EUtranCellFDD

• EUtranCellTDD

Table 11 Counters - Static Cell Connection Intensity Control


Counter Description
pmRrcConnEstabFailCe This counter is stepped at transmission of RRC Connection
llIntensStat Establishments due to max Static Cell Intensity Control
reached.
pmHoPrepRejInCellInt This counter is stepped at transmission of Handover
ensStat Preparation Failed due to max Static Cell Intensity Control
reached.
pmCellIntensHoRrcCon This counter is stepped periodically and shows the distribution
nReqDistr of the Handover Request and RRC Connection Request
intensity in relation to the Cell Connection Intensity Control limit.
pmCellIntensHoReqDis This counter is stepped periodically and shows the distribution
tr of the Handover Request intensity in relation to the Cell
Connection Intensity Control limit.

24 54/221 04-LZA 701 6014/1 Uen A1 | 2016-01-05


Performance

Counter Description
pmCellIntensRrcConnR This counter is stepped periodically and shows the distribution
eqDistr of the RRC Connection Request intensity in relation to the Cell
Connection Intensity Control limit.

More information about counters can be found in Managed Object Model


(MOM).

6.2.4 MP Load Control


Table 13 lists the counters associated with MP Load Control.

Table 13 Counters - MP Load Control


Counter Description
pmProcessorCoreLoad The average CPU load for each CPU core.
pmProcessorLoad The average CPU load on the whole MP system (all the CPU
cores).
pmProcessorLoadLcDistr A histogram counter of the MP Load value used by the MP
Load Control.
Each bin is stepped when a new CPU load sample is presented
to the MP Load Control function.

Table 14 lists the counters associated with Multiple Cell Load Modules on
multi-core DUs.

Note: The counters exist in the following MO classes:

• EUtranCellFDD

• EUtranCellTDD

Table 14 Counters - Cell allocation on multi-core DUs


Counter Description
pmRrcConnMax This counter contains the Peak number of UEs in
RRC_CONNECTED mode. The reported values for counter
pmRrcConnMax are used to set the MOM attribute
measuredMaxNoOfUsersInCell.
pmRrcConnLevSum This counter contains the sum of all sample values recorded for
"number of UEs in RRC_CONNECTED mode". The reported
values for counter pmRrcConnLevSum are used to set
the MOM attribute
measuredMaxNoOfUsersInCell.

Note: The counters pmRrcConnMax and pmRrcConnLevSum are


associated. They are not included in any predefined counter scanner.
This means that a user-defined counter scanner must be started for
these counters to get the attribute measuredMaxNoOfUsersInCell
to be updated at each ROP end.

54/221 04-LZA 701 6014/1 Uen A1 | 2016-01-05 25


Basic Load Management

More information about counters can be found in Managed Object Model


(MOM).

6.2.5 Procedure Latency Supervision


Table 17 lists the counters associated with Procedure Latency Supervision.

Note: The counters exist in the following MO classes:

• EUtranCellFDD

• EUtranCellTDD

Table 17 Counters - Procedure Latency Supervision


Counter Description
pmRrcConnEstabFailCe This counter is stepped at transmission of RRC Connection
llLatency Reject triggered by Procedure Latency Supervision.
pmHoPrepRejInCellLat This counter is stepped at transmission of Handover
ency Preparation Failed triggered by Procedure Latency Supervision.
pmRrcLatencyConnSetu This counter is stepped for each received RRC Connection
pDistr Request message and describes the procedure latency for
each message.
pmRrcLatencyConnReco This counter is stepped for each received RRC Connection
nfigDistr Reconfiguration message and describes the procedure latency
for each message.
pmRrcLatencyCapEnqDi This counter is stepped for each received RRC Capability
str Enquire message and describes the procedure latency for each
message.
pmRrcLatencySecModeC This counter is stepped for each received RRC Security Mode
mdDistr Command message and describes the procedure latency for
each message.

More information about counters can be found in Managed Object Model


(MOM).

6.2.6 Access Class Barring


For information about Access Class Barring counters, see Idle Mode Support

6.2.7 BB Management Interface Intensity Control


Table 19 lists the counters associated with BB Management Interface
Intensity Control.

Note: The counter exist in the following MO classes:

• EUtranCellFDD

• EUtranCellTDD

26 54/221 04-LZA 701 6014/1 Uen A1 | 2016-01-05


Performance

Table 19 Counters - BB Management Interface Intensity Control


Counter Description
pmRrcConnEstabFailMI This counter is stepped at transmission of RRC message RRC
SigQCong Connection Reject when RAC-BB signal queue congestion was
indicated.

More information about counters can be found in Managed Object Model


(MOM).

6.3 Events
This section describes the events for the different functions in Basic Load
Management.

6.3.1 Paging Intensity Control


This function has no associated events.

6.3.2 DU Connection Intensity Control


This function has no associated events.

6.3.3 Static Cell Connection Intensity Control


This function has no associated events.

6.3.4 MP Load Control


Table 13 lists the events associated with MP Load Control.

Table 13 Events - MP Load Control


Event Event Parameter Description
INTERNAL_PER_PR EVENT_PARAM_PROCESS Load of a specific processor
OCESSOR_LOAD OR_LOAD sampled at 10 second intervals.

54/221 04-LZA 701 6014/1 Uen A1 | 2016-01-05 27


Basic Load Management

Event Event Parameter Description


INTERNAL_EVENT_ EVENT_PARAM_ST_TIME Generated when the MP Load
LOAD_CONTROL_ST STAMP_HOUR Control state is changed due to
ATE_TRANSITION EVENT_PARAM_ST_TIME any measurement.
STAMP_MINUTE
EVENT_PARAM_ST_TIME
STAMP_SECOND
EVENT_PARAM_ST_TIME
STAMP_MILLISEC
EVENT_PARAM_TRANSIT
ION_TRIGGER
EVENT_PARAM_OLD_STA
TE
EVENT_PARAM_NEW_STA
TE
EVENT_PARAM_RBS_MOD
ULE_ID

More information about events can be found in Manage Lists.

6.3.5 Procedure Latency Supervision


Table 15 lists the events associated with Procedure Latency Supervision.

Table 15 Events - Procedure Latency Supervision


Event Event Parameter Description
INTERNAL_EVENT_ EVENT_PARAM_ADMISSI The event is generated when the
ADMISSION_BLOCK ON_ARP_LEVEL admission control function starts
ING_STARTED blocking resource requests for the
cell.
INTERNAL_EVENT_ EVENT_PARAM_ADMISSI The event is generated when the
ADMISSION_BLOCK ON_ARP_LEVEL admission control function updates
ING_UPDATED its status from Busy to Overload or
vice versa.
INTERNAL_EVENT_ N/A The event is generated when the
ADMISSION_BLOCK admission control function stops
ING_STOPPED blocking resource requests for the
cell.

More information about events can be found in Manage Lists.

6.3.6 Access Class Barring


For information about Access Class Barring events, see Idle Mode Support

6.3.7 BB Management Interface Intensity Control


This function has no associated events.

28 54/221 04-LZA 701 6014/1 Uen A1 | 2016-01-05


O&M Information

7 O&M Information

This section provides Operation and Maintenance (O&M) information for the
Basic Load Management feature.

7.1 Feature Activation


This is a Basic feature and the majority of the self contained functions are
enabled by default.

54/221 04-LZA 701 6014/1 Uen A1 | 2016-01-05 29

You might also like