Chapter 8 Creating Reports and Using Extended Accounting
Data
This chapter describes how to create reports and use extended accounting data.
The following topics are discussed:
About Reports
If you have the Performance Reporting Manager software installed, you can generate reports that
detail the CPU and the memory resources that are used per container, host, or
resource pool. For zones, you can generate reports that detail the CPU, memory, input
bandwidth, and output bandwidth. All of these graph reports are available from the
Utilization tab located in the Container Manager GUI. These graph reports become available
in the GUI two hours after the Performance Reporting Manager software is installed. This waiting period is
needed to allow data to be collected and stored in the database for use in the graph
reports. The waiting period for weekly and monthly graphs is 24 to 48 hours.
You can generate two types of utilization reports.
Table 8–1 Types of Utilization Reports
|
Resource Object
|
Total Utilization Report
|
Cumulative Utilization Report
|
|
Host
|
Utilization of all projects
|
Utilization of top 5 resource objects.
On Solaris 10, resource objects might be resource pools, zones, or projects.
On Solaris 8 and 9, resource objects might be resource pools or projects.
|
|
Pool
|
Utilization of all projects
|
Utilization of top 5 resource objects.
On Solaris 10, resource objects might be zones or projects.
On Solaris 8 and 9, resource objects are projects.
|
|
Zone
|
Utilization of all projects
|
Utilization of top 5 projects.
On Solaris 10, resource objects are projects.
|
|
Container
|
Utilization of all projects
|
Utilization of top 5 hosts.
|
|
Project
|
Utilization of the project
|
Not applicable
|
The following types of resource usage graph reports are available from
the Container Manager GUI.
- Container
-
Data returned is the average of the resources used by all containers,
which can include both active and inactive containers. Historical data is provided
for inactive containers. The addition of historical data enables you to determine
whether your containers are controlling resource consumption effectively. The data
is represented as a percentage of the resource reservations for minimum CPU and memory
cap for all active containers. This percentage compares the actual resources used
to the resources reserved.
- Active Container
-
Data returned is the number of CPUs and memory currently being used
for the selected active container.
- Container Group
-
Data returned is the average of the resource used for all containers
in the selected group. This percentage compares the actual resources used to the resources
reserved for the selected containers.
- Host
-
Data returned is the aggregation of all active containers on the
selected host.
- Host Group
-
Data returned is the average resource utilization of all hosts located
in that group. The data is represented as a percentage used of the total host resources.
- Resource Pool
-
Data returned is the aggregation of all the active containers in
the selected resource pool.
- Zone
-
Data returned is the aggregation of all the active projects in the
selected zone.
If the requested graph is for multiple containers across different hosts,
the data returned is the average of the percentage being used on each host.
Report data can also be exported to a text file in comma-separated values
(CSV) format for an active container, zones, resource pool, or host. The text file
can be used as an interface file for a billing and accounting application, for example.
A report in CSV format is available 2 hours after installation of the Performance Reporting Manager software.
This waiting period enables report data to be collected and stored in the database
for use in a CSV report. The exported data is more detailed and granular than the
data that appears in the graph reports. Data for the last 24 hours is available in
a CSV report.
The exported CSV reports contains the following categories of information:
- Host name
-
Name of the host with which the container is associated
- Timestamp
-
Date and time for the record.
- CPU Reservation
-
CPU reservation of the container
- CPU Usage
-
Combined CPU usage of all processes in the container
- CPU Return of Investment
-
CPU utilization compared to CPU reserved, expressed as a percentage.
- CPU Extended Accounting Information
-
CPU extended accounting information
- Memory Cap
-
Physical memory cap
- Memory Usage
-
Physical memory used
- Percentage of Memory Used
-
Physical memory utilized of the host expressed as a percentage
- Memory Return of Investment
-
Memory utilized compared to memory reserved, expressed as a percentage.
- Container Project ID
-
Project ID of the container
- Project Name
-
Project name of the container
Data Collection Process
Container Manager uses the Performance Reporting Manager data collection service, which is located on the server
layer. This data collection service in turn uses the history logging capabilities
of Sun Management Center, which is located on the agent layer. The data collection service on the
server layer collects the data from the agent machines and stores it in the database.
Additionally, data collected by Performance Reporting Manager is summarized, or “rolled-up,” at
predefined intervals. The minimum, maximum, and average values for data are calculated
and are stored as hourly data, weekly data, or monthly data. Finally, the minimum,
maximum, and average values are calculated for these same intervals.
The reports generated with Container Manager can incorporate any of this data, depending
upon the report request parameters. For more information about Performance Reporting Manager data collection
methods, see “Data Collection Process” in Sun Management Center
3.6.1 Performance Reporting Manager User’s Guide.
Requesting a Report
Both CPU and memory resource utilization reports are available forhost, host group, container, container group,
resource pool, zone, and active container. Before trying to view a report, be sure
to set the browser's cache to refresh every time. For a list of the six types of reports
available, see About Reports. Reports for
CPU and memory resources used are available for the following intervals:
-
Daily – Shows data for the last 24 hours, one point per hour
-
Weekly – Shows data for the last 7 days, one point per day
-
Monthly – Shows data for the last 30 days, one point per day
-
Real Time - Shows data every 30 seconds per graph point. Available
for active projects only.
You must wait two hours after installation of the Performance Reporting Manager software for the daily
graph reports to become available. Data first must be collected and stored in the
database from which the report can be drawn. You can also view CPU and memory resource
utilization reports for inactive containers and containers that are based on historical
data.
Real time reports for CPU and memory resources being used are available for
active projects only.
To Request a Resource Utilization Report For
a Host
Use this procedure if you want to obtain a daily, weekly, or monthly report
for a host.
Steps
-
Set the browser's cache to refresh every time.
-
In the Hosts view, select the host from the navigation window.
-
Select the Utilization tab.
The CPU
usage and memory usage of the host for the last 24 hours are displayed.
Note –
(On Solaris 10) The CPU usage and the memory usage are displayed.
-
Click the Graph Properties link.
-
Choose whether to generate a report for the entire host or only for a
resource object in the host.
Note –
The resource object for a host might be a resource pool or a project.
(On Solaris 10) The resource object for a host might be a resource pool,
zone, or a project.
-
To generate a report for the entire host, select the Total Utilization
option.
Select the appropriate time interval from the Time Interval list.
Click the Update Graph button.
The CPU and memory resource
utilization graphs appear.
-
To generate a report for a resource object in the host, select the
Cumulative Utilization by Object option.
Select the appropriate resource
object.
Select the appropriate time interval from the Time Interval list.
Click the Update Graph button.
The CPU and memory resource
utilization graphs appear for a maximum of 5 resource objects that have high utilization.
-
(Optional) To export the last 24 hours of data
to a CSV file, click the Export Data button.
You must wait at least 2
hours after installation of the Performance Reporting Manager software for a CSV report to be available. Data
must first be collected and stored in the database from which the report can be drawn.
You cannot preview this data in a graph.
Data exported contains the hourly
data for the container for the last 24 hours. Therefore, it is not identical to
the data obtained from a daily graph.
To Request a Resource Utilization Report for
an Active Project
Use this procedure if you want to obtain a daily, weekly, monthly, or real time
report for an active project.
Steps
-
Set the browser's cache to refresh every time.
-
In the Hosts view, select the host from the navigation window.
-
Select the Projects tab in the right pane.
A list of projects on this host appears.
-
Select the active project for which you need
a report.
-
Select the Utilization tab.
The CPU usage and memory usage
of the active project for the last 24 hours are displayed.
-
Click the Graph Properties link.
-
Select the Total Utilization option.
-
Select the appropriate time interval from the Time Interval list.
You can also obtain a real time report for an active project.
-
Click the Update Graph button.
The CPU and memory resource
utilization graphs appear.
-
(Optional) To export the last 24 hours of data
to a CSV file, click the Export Data button.
You must wait at least 2
hours after installation of the Performance Reporting Manager software for a CSV report to be available. Data
must first be collected and stored in the database from which the report can be drawn.
You cannot preview this data in a graph.
Data exported contains the hourly
data for the container for the last 24 hours. Therefore, it is not identical to
the data obtained from a daily graph.
To Request a Resource Utilization Report for
a Container
Use this procedure to request CPU and memory utilization reports for a container.
The data is based on historical data and is an average of the resources used by active containers
that are based on the container.
Steps
-
Set the browser's cache to refresh every time.
-
In the Containers view, select the container.
-
Select the Utilization tab.
The CPU usage and memory
usage of the container for the last 24 hours are displayed.
-
Click the Graph Properties link.
-
Choose whether to generate a report for the entire container or for a
resource object in the container.
Note –
The resource object for a container is a host.
-
To generate a report for the entire container, select the Total Utilization
option.
Select the appropriate time interval from the Time Interval list.
Click the Update Graph button.
The CPU and memory resource
utilization graphs appear.
-
To generate a report for a resource object in the container, select
the Cumulative Utilization by Object option.
Select the appropriate resource
object.
Select the appropriate time interval from the Time Interval list.
Click the Update Graph button.
The CPU and memory resource
utilization graphs appear for a maximum of 5 resource objects that have high utilization.
To Request a Resource Utilization Report for
a Resource Pool
Use this procedure if you want to obtain a daily, weekly, or monthly report
for a resource pool.
Steps
-
Set the browser's cache to refresh every time.
-
In the Hosts view, select the host to which the resource pool is bound.
A list of all resource pools bound to this host appears in the Resource
Pools table in the right pane.
-
Select the name of the resource pool in the table.
A table listing all containers that are bound to this resource pool appears.
-
Select the Utilization tab.
The CPU
usage and memory usage of the resource pool for the last 24 hours are displayed.
Note –
(On Solaris 10) The CPU usage and memory usage are displayed.
-
Click the Graph Properties link.
-
Choose whether to generate a report for the entire resource pool or for
a resource object in the resource pool.
Note –
The resource object for a resource pool is a project.
(On
Solaris 10) The resource object for a resource pool will be zones and projects.
-
To generate a report for the entire resource pool, select the Total
Utilization option.
Select the appropriate time interval from the Time
Interval list.
Click the Update Graph button.
The CPU and
memory resource utilization graphs appear. Also, utilization report and bandwidth
reports are generated.
-
To generate a report for a resource object in the resource pool, select
the Cumulative Utilization by Object option.
Select the appropriate resource
object.
Select the appropriate time interval from the Time Interval list.
Click the Update Graph button.
The CPU and memory resource
utilization graphs appear for a maximum of 5 resource objects that have high utilization.
Also, utilization report and bandwidth reports are generated.
-
(Optional) To export the last 24 hours of data
to a CSV file, click the Export Data button.
You must wait at least 2
hours after installation of the Performance Reporting Manager software for a CSV report to be available. Data
must first be collected and stored in the database from which the report can draw
from. You cannot preview this data in a graph.
Data exported contains
the hourly data for the container for the last 24 hours. Therefore, it is not identical
to the data obtained from a daily graph.
(Solaris 10 Only) To Request a Resource Utilization Report
For a Zone
Use this procedure if you want to obtain a daily, weekly, or monthly report
for a zone.
Steps
-
Set the browser's cache to refresh every time.
-
In the Hosts view, select the Solaris 10 host from the navigation window.
-
Select the Zones tab.
A list of zones on this host
appears.
-
Select a zone for which you need a report.
-
Select the Utilization tab.
The CPU usage, memory usage, and
bandwidth usage of the zone for the last 24 hours are displayed.
-
Choose whether to generate a report for the entire zone or only for a
resource object in the zone.
Note –
The resource object for a zone is a project.
-
To generate a report for the entire zone, select the Total Utilization
option.
Select the appropriate time interval from the Time Interval list.
Click the Update Graph button.
The CPU and memory resource
utilization graphs appear.
-
To generate a report for a resource object in the zone, select the
Cumulative Utilization by Object option.
Select Projects in the list.
Select the appropriate time interval from the Time Interval list.
Click the Update Graph button.
The CPU and memory resource utilization
graphs appear for a maximum of 5 resource objects that have high utilization.
-
(Optional) To export the last 24 hours of data to a CSV file, click the
Export Data button.
You must wait at least 2 hours after installation
of the Performance Reporting Manager software for a CSV report to be available. Data must first be collected
and stored in the database from which the report can draw from. You cannot preview
this data in a graph.
Data exported contains the hourly data for the container for
the last 24 hours. Therefore, it is not identical to the data obtained from a daily
graph.
Ladda ner denna bok i PDF (1430 KB)