Data storage in time series database

Database schema that stores the performance metric data that is collected with the help of Timeseries Service. The time series data consists of both performance raw metrics, aggregated metrics, UDCs, and metric threshold values.

Metric types

The following types of metrics are stored in time series database:

Raw metrics
A raw count of some network performance event.
User-Defined Calculations
A mathematical expression based on one or more raw metrics, or other User-Defined Calculations that are available within the Telco Network Cloud Manager - Performance system. They are also called as complex metrics. You can either create User-Defined Calculations from User-Defined Calculations configuration UI or available in some Technology Packs. See Managing User-defined calculations (UDC).
Analytic metrics
Aggregated metrics that are analyzed from the configured batch jobs, streams, and stored busy hours. These metrics are used in Telco Network Cloud Manager - Performance Dashboards and system configuration pages.

Streams can be created from Streaming Analytics configuration page or available in some Technology Packs. See Managing streams for metrics.

Metric origin

Metrics that are stored in Timeseries Service have the origin property to differentiate from where the metric is coming from. The different types of origin based on the type of metric that are being saved are as follows:

Metrics that come from installed Technology Packs do not have the origin property.

For example,


name=CPU.Utilization.Percent
description="The percentage of CPU utilization"
units=Percent
aliases=[cpuBusyPoll,JuniperERX CPU Utilization,JuniperChassis CPU Utilization avg last 5min,Huawei CPU Utilization,CiscoDevice System CPU Utilization CPM]
properties={
             resource-types="cpu,card,device",
             baselineEnabled="true"
}

Note: The CPU-Utilization-Percent.metric is baseline-enabled metric as you can see from the definition.

Metrics that are configured as batch jobs, streams, and busy hours have the origin as Analytic.

For example, for a metric that is configured for batch analytics, tavg_15min_CPU.Utilization.Percent.metric.


name="tavg_15min_CPU.Utilization.Percent"
description="Metric created from Batch Analytics"
units=""
aliases=[]
properties={
origin="Analytic"
aggregator="Average"
resource-types="device"
}

For example, for a metric that is configured for streaming analytics, ssum_Region.Huawei_2G_Call_Drop_Rate_Percent.metric.


name="ssum_Region.Huawei_2G_Call_Drop_Rate_Percent"
description="Metric created from Streaming Analytics"
units=""
aliases=[]
properties={
"resource-types":"region",
"aggregator":"Sum",
"origin":"Analytic"
}

For example, for a metric that is configured for Busy Hour, sbhd_min_day_nUtranCell.N.CA.SCell.Add.Att.metric.

name="sbhd_min_day_nUtranCell.N.CA.SCell.Add.Att"
description="Busy Hour Determiner"
units=""
aliases=[]
properties":{
   "resource-types":"nUtranCell",
   "aggregator":"min",
   "origin":"Analytic"
}

Metrics that are configured as UDCs, which can also be called as complex metrics have the origin property as UserDefined.
For example, for a metric that is configured as UDC, Region.Huawei_3G_Call_Drop_Rate_CS_Percent.metric.
```
name="Region.Huawei_3G_Call_Drop_Rate_CS_Percent"
description="Call drop rate"
units="percent"
aliases=[]
properties={
origin="UserDefined"
aggregator="Sum"
resource-types="region"
}
```