Contents1 Introduction
This document describes the performance indicators for the Virtual Multimedia Resource Function (vMRF).
1.1 Scope
This document describes how the vMRF counters can be used for network planning and dimensioning of resources.
The document describes vMRF measurements divided as follows:
The indicators are grouped into the following categories:
-
Accessibility
-
Retainability
-
Integrity
-
Traffic and load
-
Compute resource
2 vMRF Performance Management Overview
vMRF performance indicators are of the following types:
| Key Performance Indicators (KPIs) |
KPIs are the main measurement indicators, used to highlight the status of vital functionality. |
|
| Performance Indicators (PIs) |
PIs are supporting measurement indicators that provide additional information for all functionality. |
|
The indicators are grouped into the following categories:
| Accessibility |
The ability to obtain requested services from the system to a GCP (H.248) termination |
|
| Retainability |
The ability for a GCP (H.248) context to retain its requested services, once connected, for the desired duration |
|
| Integrity |
The ability of an external connection to maintain the connection at desired quality |
|
| Traffic and load |
This category provides information about the current status of a node, mainly from resource use point of view. |
|
| Compute resource |
This category provides information about the current status of a node, from compute resource use point of view. |
|
For more information on measurement and performance management, refer to Performance Management.
3 Resource Service KPIs in vMRF
Table 1 shows the Resource Service Key Performance Indicators (R-KPIs) supported in vMRF.
|
Category |
KPI Name |
|
|---|---|---|
|
Accessibility |
||
|
Retainability |
||
3.1 KPIs and PIs for Accessibility
3.1.1 Session Setup Success Ratio KPI
The Session Setup Success Ratio is used to measure the ratio of successful call setup initiations. Session setups are considered successful even if the call is not answered or is rejected by the terminating user or if a fault outside the IMS network arises.
| Note: |
In the vMRF, Session Setup Success Ratio measures the session setup from receiving the H.248 Add request to sending the H.248 Add reply message. |
Formula for the measurement: (1 – ([for all MrfH248Interface].terminationReqsRej / [for all MrfH248Interface].terminationReqs)) * 100
Target value range: 99.7–100% (for long term measurement)
Possible reasons for falling below the target value range:
-
Congestion in one or more MRF resources
-
IPv4 connectivity is faulty. If the VM is configured exclusively for IPv4 sessions, all sessions are rejected. If the VM is configured for both IPv4 and IPv6 sessions, only IPv4 sessions are rejected.
-
IPv6 connectivity is faulty. If the VM is configured exclusively for IPv6 sessions, all sessions are rejected. If the VM is configured for both IPv4 and IPv6 sessions, only IPv6 sessions are rejected.
Possible consequence of falling below the target value range:
-
Increased traffic rejection rate
Recommended actions when falling below the target value range:
-
Check the status of related resources and devices.
3.2 KPIs and PIs for Retainability
3.2.1 Session Completion Ratio KPI
The Session Completion Ratio shows the ability of the vMRF to successfully complete an active session in the IMS network. Sessions are considered successfully completed even if a fault arises outside the IMS network or when the session is terminated due to the lack of support for held SRVCC calls in the IMS.
Formula for the measurement: (1 – ([for all MrfH248Interface].abnormTermTerminations / ([for all MrfH248Interface].terminationReqs – [for all MrfH248Interface].terminationReqsRej))) * 100
Target value range: 99.6–100% (for long term measurement)
Possible reasons for falling below the target value range:
-
Problems in the network
-
Faults in MRF resources
-
Forced locking of MrfInstance MOs
Possible consequence of falling below the healthy value range:
-
Increased amount of dropped calls
Recommended action when falling below the healthy value range:
-
Check the status of related resources and devices.
4 Service Level Agreement PIs in vMRF
Table 2 shows the Service Level Agreement Performance Indicators (SLA-PIs) supported in vMRF. These PIs can be used to monitor if the minimum hardware requirements set by a VNF are fulfilled by the cloud environment. This information is needed for troubleshooting, to detect if the source of the problem lies in the VNF, in the cloud environment, or in the external network.
|
Category |
PI Name |
|---|---|
|
Compute resource |
|
4.1 CPU Steal Time PI
The CPU Steal Time PI is used to monitor if CPU pinning is used. If the PM threshold value is exceeded, the Service Level Agreement Violation alarm is raised.
Threshold value: 3%.
Possible consequence of exceeding the threshold value:
-
Negative impact on QoS related to vMRF services.
-
Contact the Cloud Administrator to allocate the required amount of resources for the vMRF, as described in vMRF Infrastructure Requirements.
4.2 Allocated Memory PI
The Allocated Memory PI is used to monitor if the minimum memory requirements for a VM are fulfilled. If the PM threshold value is exceeded, the Service Level Agreement Violation alarm is raised.
Threshold value: 80%
-
The vMRF VM allocated too little memory.
Possible consequence of exceeding the threshold value:
-
Call failures can occur if memory use further increases.
-
Contact the Cloud Administrator to allocate the required amount of resources for the vMRF, as described in vMRF Infrastructure Requirements.
4.3 Swap Memory PI
The Swap Memory PI is used to monitor if memory swap occurs. If the PM threshold value is exceeded, the Service Level Agreement Violation alarm is raised.
Threshold value: 3%
-
The vMRF VM allocated too little memory.
Possible consequence of exceeding the threshold value:
-
Negative impact on QoS related to vMRF services.
-
Contact the Cloud Administrator to allocate the required amount of resources for the vMRF, as described in vMRF Infrastructure Requirements.
4.4 Allocated Disk PI
The Allocated Disk PI is used to monitor if minimum disk requirements for a VM are fulfilled. If the PM threshold value is exceeded, the Service Level Agreement Violation alarm is raised.
Threshold value: 50%
Possible consequence of exceeding the threshold value:
-
No traffic impact, storage of new files on the disk might fail.
-
Allocate more storage space to meet requirements, as described in vMRF Infrastructure Requirements.
-
Clean up disk space by removing possible core dumps from /cluster/storage/dumps/ and tracing-related files from /cluster/storage/collectdata.