Capacity Monitoring: Overview

You can view the aggregated utilization trends for resources or group of resources in data centers in the Power® environment. You can also monitor data center efficiency by using energy metrics.

To collect the performance data, you must enable the data collection for the managed systems in the management console where the cloud connector is running. For more information about enabling data collection, see Enabling data collection. For information about enabling metrics collection for SSP, see SSP Preferences.

You must enable the NTP service on the HMC (Hardware Management Console) to automatically synchronize the HMC clock with an NTP time server. When you enable the service, the time of your HMC is synchronized with other management consoles. The applications that are running in the cloud depend on the data that is captured in HMC that has local HMC time stamp. If you do not configure the NTP service in the HMC, there can be inconsistency in the time at which the data is collected by the Cloud Connector in the management console and the view that is built in the cloud application might not show the real-time data. To enable the NTP service, log in to the HMC and go to the HMC Management > Change date and time> NTP configuration and select the Enable NTP service for this HMCcheck-box.

Before you can view the utilization trend for the selected resource or group of resources, you must select the chart type and select specific resources or resource group. The Charts area displays the resource types for which you can view the performance data as graphs. Select a predefined time interval for which you want to view the utilization trend from the Time Range list. In the Resources area, you can search for the specific resource or a resource group by entering the text string in the search field for the selected resource type. The search result lists resources and groups that match the search criteria. From the list, select the resources or a resource group for which you want to view the utilization trend and add it to the resources list. Click Apply to apply your selections. The graphs that are associated with the selected resource type are displayed in the dashboard. By default, the utilization data for the last 2 hours is displayed for the selected resources.

You can also specify the date and time range for which you want to view the utilization trend by selecting Custom Range from the Time Range list. Depending on the selected time interval, the aggregated utilization data is displayed for the selected resource.

You can modify the resource selection by adding more resources to the list or by removing resources from the list, and then click Apply to apply the new search criteria. You can save the configured search criteria to reuse the search criteria later. To save the search criteria, click Save Search. Specify a name for the criteria and click OK. The saved search is listed in the Saved Searches list. You can also edit the saved search and save it again.

Select Data auto-refresh from the navigation area to automatically refresh the data in the graphs. You can also select the refresh time to automatically update the graphs. Select the required time from the Refresh Interval list to automatically update the graph values every 30 seconds, 1 minute, 2 minutes, or 5 minutes. You cannot select the refresh interval when you select Custom Range from the Time Range list.

The following table lists the graphs that are associated with each resource view and the attributes that are used to plot the graphs.

Table 1. Utilization graphs

Resource view	Associated graphs	Attributes
managed system	CPU utilization	utilized processing units
	assigned LPAR memory utilization	assigned memory to LPAR
	network utilization sent bytes received bytes	all VIOS generic network adapters all SRIOV physical ports all HEA Physical Ports
	storage utilization read bytes write bytes	all VIOS generic physical adapters all VIOS Fibre Channel adapters
	shared processor pool utilization
	system power utilization
logical partition and Virtual I/O Server (VIOS)	CPU utilization	utilized processing units
	memory utilization	VIOS - utilized memory logical partition - backed physical memory
	network utilization sent bytes received bytes	VIOS VIOS generic network adapters SRIOV logical ports logical partition virtual Ethernet adapters SRIOV logical ports
	storage utilization read bytes write bytes	VIOS generic physical adapters Fibre Channel adapters logical partition generic virtual adapters virtual Fibre Channel adapters
shared storage pool (SSP)	utilization at SSP level read bytes write bytes
	utilization per node level read bytes write bytes
	utilization per tier at each node level read bytes write bytes

In the management console, the processed metrics for a managed system, logical partition, and VIOS are generated every 30 seconds. The aggregated metrics are generated every 5 minutes, 2 hours and 24 hours. The graphs are plotted with processed metrics or aggregated metrics based on the following conditions:

• If the selected time interval is less than 1 hour, then processed metrics are used.
• If the selected time interval is between 1 hour and 12 hours, then 5 minutes sample is used.
• If the selected time interval is between 12 hours and 7 days, then 2 hours sample is used.
• If the selected time interval is more than 7 days, then 24 hours sample is used.

For SSP, the processed metrics are generated every 5 minutes. The aggregated metrics are generated every 2 hours and 24 hours. The graphs are plotted depending on the following conditions:

• If the selected time interval is less than 12 hours, then 5 minutes sample is used.
• If the selected time interval is between 12 hours and 7 days, then 2 hours sample is used.
• If the selected time interval is more than 7 days, then 24 hours sample is used.

If some samples are not available, the graphs are plotted with the interpolated data.