Configuring and managing thresholds for metrics and indicators

Use the Thresholds page in the TrueSight Capacity Optimization console to configure and manage thresholds for metrics and indicators.

To access the Thresholds page, click Administration > OPTIMIZER > Thresholds. The page has two tabs:

  • Metrics: Lists the metric thresholds and enables you to manage them. With thresholds, you can monitor how the metric is behaving relative to a specified value. You can define Good and Warning thresholds for a metric. When a Warning threshold is breached for a specified period, the Optimizer rule that uses the metric threshold generates an alert. You can configure the scope of a metric threshold at two levels: 
    • Global: Threshold applies to the metric on all systems or all business drivers 
    • System-specific: Threshold applies to the metric on one of the following levels:
      • Specific system or business driver
      • All systems or business drivers of a specific type
      • Systems or business drivers in all domains
      • All systems or business drivers of a specific domain
    You can create multiple thresholds for the same metric and define a different scope as per your requirement. If both Global and System-specific thresholds are set for a metric, the System-specific threshold setting overrides the Global threshold setting. 
    Some default thresholds are provided out-of-the-box. You can add new ones, or modify and delete the existing thresholds from the Metric thresholds page. For more information, see To configure and manage metric thresholds.
  • Indicators: Lists the indicator parameters and thresholds and enables you to reconfigure them as per your requirements. These indicator parameters and thresholds are used to compute indicator values. The default parameter and threshold values are provided out-of-the-box. You cannot add new indicator thresholds or delete existing ones. For more information, see To configure indicator parameters and thresholds

To configure and manage metric thresholds

To view metric thresholds

Click Administration > OPTIMIZER > Thresholds. Ensure that the Metrics tab is selected.

All the metric thresholds are displayed in a table. Here are the column descriptions:


Column

Description

Entity iconIcon that depicts the type of entity to which the metric belongs. For more information, see Entity types.

Metric

Name of the metric for which the threshold is defined.

Scope

Specifies whether the threshold applies to the metric across all or specific entities (systems or business drivers). One of the following values is displayed:

  • Global - Threshold applies to the metric in all systems or business drivers
  • System specific - Threshold applies to the metric in specific systems or business drivers depending on the configuration.

A metric can have multiple records in the table depending on its scope.

Note

If both Global and System specific thresholds are set for a metric, the System specific threshold setting overrides the Global threshold setting.

For example: Consider that thresholds with the following properties are configured for the CPU Utilization % metric:


ScopeSystem/Business driver typeDomainGood thresholdWarn threshold
1Globalanyany65%95%
2Domain specificanyAIX Power VM/WPAR75%90%

When applying thresholds for the CPU Utilization % metric on a VMware system, the first one (Global) is used. However, if the system belongs to the AIX Power VM/WPAR domain, then the second threshold (Domain specific) is applied.


System/business driver type

Type of system on which the threshold is to be set, for example, KVM host, Solaris.

System/business driver name

Name of the system or business driver to which the metric threshold is applicable. If the scope of the threshold is Global, then any is displayed.

Domain name

Name of the domain of the system or business driver. If no specific domain is selected during the threshold configuration, then All domains is displayed.

Good Threshold

Good threshold value that is defined for the metric.

Warn Threshold

Warning threshold value that is defined for the metric. If this value is arrived at, or exceeded, you will be warned.

Limit Resource

Metric on which the threshold is based. If a metric threshold is based on a fixed value, then this field is blank.

To add a metric threshold

  1. Click Administration > OPTIMIZER > Threshold. By default, the Metric thresholds page is displayed.
  2. Click Add Threshold.
  3. Specify the following details:
    1. In Select structure, select whether you want to specify thresholds for systems or business drivers. Depending on the option that you select in this step, the next option changes.
    2. From Select system or Select business driver, select whether the thresholds must apply to all or specific systems or business drivers. If you select All systems or All business drivers, then the scope of the metric threshold is set to be global, else the scope depends on the specific system or business driver that you specify.

    3. In Metric, select the metric for which you want to add threshold values.
    4. In Domain, select whether to apply the metric threshold to systems or business drivers in all domains or to a selected domain.
    5. In Value, select whether you want to specify fixed values for good or warning thresholds or whether you want to base the thresholds on another metric.
      • If you select Fixed value, specify a fixed Good Threshold and Warn Threshold, measured in Operations/second.
      • If you select Based on limit metric, select a metric and specify a Good Threshold and Warn Threshold, measured in percentage of the limit metric.

      Note

      A metric value is evaluated against the thresholds as follows:

      • Good: If the metric value is less than the 'Good threshold.'
      • Warn: If the metric value lies between the 'Good threshold' and 'Warn threshold.'
      • Poor or Critical: If the metric value is greater than the 'Warn threshold.'
  4. Click Save.
    The threshold is added to the table on the Metric thresholds page.

To edit a metric threshold

  1. Click Administration > OPTIMIZER > Threshold. The Metric thresholds page is displayed by default.
  2. In the table on the page, for the threshold that you want to modify, click the Edit icon.
  3. In the Edit threshold section of the page, modify the Good Threshold and Warn Threshold values.
  4. Click Save.

To delete a metric threshold

  1. Click Administration > OPTIMIZER > Threshold. The Metric thresholds page is displayed by default.
  2. In the table on the page, for the threshold that you want to delete, click the Delete icon.
  3. Click Delete.

To configure indicator parameters and thresholds

Access the Indicators page by clicking Administration > OPTIMIZER > Thresholds > Indicators. The indicator parameters and thresholds are displayed.

To modify these values, scroll down to the bottom of the page and click Edit, and after completing the changes, click Save. You can neither add nor delete a parameter or threshold.

The following table provides a list of the indicator parameters and thresholds, their default values, and the indicator that is impacted when their values change. For more information about indicators, see Indicators.

Indicator parameter or thresholdDefault value Impacted indicator
Days to saturation
Minimum number of days needed to calculate trends5 days

Days to saturation

(IND_DAYS_TO_SATURATION)

Idle VMs

CPU utilization threshold5 %
  • Hours idle (IND_HOURS_IDLE)
  • Days idle (IND_DAYS_IDLE)


Network rate threshold5 Mbytes/s
Disk I/O rate threshold5 Mbytes/s
Overallocated VMs
vSphere VM Active Memory increase factor50%

Memory Demand

(IND_MEMORY_DEMAND)

Virtualization Readiness

CPU consumption metric mandatory

True

Virtualization Readiness Index

(IND_VIRT_READINESS)



CPU consumption threshold

5 GHz

Memory consumption metric mandatory

True

Memory consumption threshold

8 GB

Disk consumption metric mandatory

False

Disk consumption threshold

7.5 Kbyte/s

Network consumption metric mandatory

False

Network consumption threshold

1 Gbit/s

CPU/Memory/Storage Pressure
CPU Pressure Indicator for VMware Clusters and Hosts
CPU Ready threshold for VMs10%

CPU Pressure Index

(IND_CPU_PRESSURE)

Minimum VMs considered for CPU Ready impact10
Composite Memory (Memory Pressure Indicator) for VMware Clusters, Hosts and Capacity Pools
Ratio of consumed memory by active memory threshold6
  • Composite Memory Consumption
    (IND_COMPOSITE_MEMORY_CONSUMPTION)

  • Composite Memory Utilization %
    (IND_COMPOSITE_MEMORY_UTILIZATION)

  • Memory Pressure Index
    (IND_MEMORY_PRESSURE)
Memory overcommitment threshold1.2
Storage Pressure Index for VMware Clusters and Capacity Pools 
I/O latency threshold30 MILLIS

Storage Pressure Index

(IND_STORAGE_PRESSURE)

Minimum Datastores that can exceed I/O latency threshold10

CPU Pressure Indicator for AIX PowerVM Hosts

CPU Queue Length threshold for partitions4

CPU Pressure Index

(IND_CPU_PRESSURE)

Minimum Partitions considered for CPU Queue Length impact5

Memory Pressure Indicator for AIX PowerVM Hosts

Memory page rate threshold for partitions64 bytes/s

Memory Pressure Index

(IND_MEMORY_PRESSURE)

Minimum Partitions considered for Memory page rate impact20

Memory Pressure Indicator for Virtual Machines

Minimum swap rate exceed count1

Memory Pressure Index

(IND_MEMORY_PRESSURE)

Spare VMs

Use Composite Memory Utilization indicator instead of Memory Utilization metric

True

Spare VMs

(IND_SPARE_VM)

Default VM CPU usage

1 GHz

Default VM memory usage

2 GB

Default VM storage usage

16 GB

Was this page helpful? Yes No Submitting... Thank you

Comments