Computer Performance reports are used to measure server performance such as CPU, Memory usage, Disk and Network performance.
This report category also includes the Performance Dashboard, which is described under Dashboards category.
The Overview report gives more detailed information for the selected computer. Values and charts in the Overview report are updated every 60 seconds. The charts display a performance overview for the last 15 minutes, while the value next to a chart is in real time. The Overview report gives you more detailed information about all Hard Drives and Network Adapters on your computer. Performance Counters available on this report are similar to those in the previously mentioned Dashboard with the addition of Computer Info and Top 5 processes categories.
This report allows you to open a Remote Desktop connection to the selected computer.
The Top 5 processes category displays CPU and Memory consumption charts for the top five Windows processes according to CPU usage. This category will quickly show you which processes are consuming the most CPU time on the selected computer. It also displays process performance counters:
IO Operations/sec and the total IOs/sec (IO Read and Write KB/sec).
Values and charts on the Overview report change colors depending on the defined thresholds for the selected computers, which are configured through the Monitoring Templates category on the Administration tab. When a chart is painted red (or yellow), it means that a threshold has been crossed for the associated performance counter. Red indicates a critical threshold, while yellow indicates a warning threshold.
Every chart and value in this report is clickable, which makes it very useful as an initial step in identifying performance bottlenecks and for further detailed analysis. When you click on a chart, the resulting drilldown will capture the Filter values and take you to a more detailed history report.
This report displays data for a single computer only.
Depending on whether the custom perfromance counters and important services are configured for the selected computer, there is one more category that can be visible here: Monitoring Templates.
Depending on what is configured through the assigned Monitoring Template, this category can contain two subcategories: Performance Counters and Monitored Services.
Defined custom performance counters will be shown with charts and values, like other (built-in) performance counters. Every custom performance counter configured for monitoring will be displayed with its category name and instance name.
You can create separate Monitoring Templates for Windows Services, assign them to designated computers or computer groups according to your preferences and define actions for stopped services. Upon configuration, the Monitoring Templates category will appear on the Overview report.
Defined Windows Services will be listed with the following columns:
- Name – The display name of the service.
- Start Mode – The startup type of the service.
- Logon account – The logon account column shows which user account the service is using to log on.
- Status – The service status column, shows the current status of all monitored services configured to run on the selected computer and is updated every 30 seconds.
If any of monitored services has the service status
Stopped, this means, depending on the defined actions, that SysKit has tried to restart or is in the process of restarting the stopped service(s) on the selected computer. Depending on the options you configure in the Monitoring Template which is assigned to a computer, SysKit can send you e-mail notifications when one or more important services are stopped and can perform automatic service restarts. It will also tell you whether it has been successful or not.
See the Monitoring Templates article to learn more about managing custom performance counters and important services.
CPU, Memory, Disk and Network
The CPU report in the Computer Performance category displays detailed CPU utilization statistics for the selected Date Range and Computer(s).
The Memory report in the Computer Performance category displays detailed Memory usage statistics for the selected Date Range and Computer(s).
The Disk report in the Computer Performance category displays detailed Disk performance statistics for the selected Date Range, Computer(s), Disk(s), and Disk Counter(s). With these two additional filters, Disks and Disk Counters, you can select the desired Logical Disk and Disk Counter for the computer(s) specified in the Computers filter.
Disk Counters explanation
- Used Disk Space – This counter displays the amount of available space and the total usable space on the logical volume in GB.
- Avg. Disk Queue Length – Average number of both read and write requests that were queued for the disk, value should be between 0 and 2 most of the time, but if you have SAN or RAID controller this can spike above 2.
- Disk Transfers/sec – Number of read and write IOPS (Input/output Operations per Second), mechanical disk IOPS are from 100-400, SSD from few thousands up to 100 000, modern SAN from few thousands up to million IOPS.
- Disk Read KB/s – Current disk read bytes per second (throughput), maximum mechanical disk rate is 50-200 MB/s, SSD up to 550 MB/s, modern SAN more than 1000 MB/s, this transfer speed also depends on the RAID level, caching mechanism etc.
- Disk Write KB/s – Average disk write bytes per second (throughput), maximum mechanical disk rate is 40-180 MB/s, SSD up to 500 MB/s, modern SAN more than 1000 MB/s, this transfer speed also depends on the RAID level, caching mechanism etc.
The Network report in the Computer Performance category displays detailed Network performance statistics for the selected Date Range, Computer(s), Network Adapter(s), and Network Counter(s). With these two additional filters, Network Adapters and Network Counters, you can select the desired Network Adapter and Network Counter for the computer(s) specified in the Computers filter.
Network Counters explanation
- Received Mbps – This counter shows the rate at which bytes are received per second over each network adapter.
- Sent Mbps – This counter shows the rate at which bytes are sent per second over each network adapter.
These reports include a few tricks:
- You can quickly zoom in and out of a chart’s diagram using the mouse wheel.
- When you click on the Drill To Applications button, the resulting drilldown will capture the Filter values and take you to a more detailed Application Performance – History report.
A small but very important item to note here is the tooltip. The tooltip displays details about the data point the mouse is currently hovering over, such as the computer name, date and time values, and value of the data point.
Use the Date Range filter to set date boundaries on your performance data. You can select any day, week, or month or a custom date range.
You will not be able see any performance data outside 30 days unless you change the Performance Counters Data Retention settings. The default is set to delete performance counters data older than 30 days.
Depending on whether the report type is real-time or history, you will have a different set of options and filters available. Use the reports ribbon to take the following actions:
- Export – Save the current report to PDF, Excel, HTML, or CSV files.
- Subscription Manager – Create, edit, and send subscriptions that contain reports and views. Subscriptions can be delivered to email, SharePoint, and FileShare.
- Schedule This Report – Schedule the current report or view to email, SharePoint, or FileShare.
- Save As View – Each of the reports in this group can have different filters applied – Computers, Disks, Disk Counters, Network Adapters, or Network Counters. To create a View, click on the Save As View button on the ribbon, enter the view name, choose the view type and visibility, and click on Save. Afterwards, you can adjust the filters on the newly created view, and save the changes by clicking the Save View Changes button on the ribbon.
- Drill To Applications – Easily drill down from the reports to Application Performance – History by clicking the Drill To Applications button on the ribbon.
- Refresh – Refresh items in the left navigation panel, filters, and the main view.
- Options – Allows configuration of options for Report and Dashboard data, Alerts, Export and System Jobs.
The Detailed Analysis report displays the history data for monitored custom performance counters, which are assigned to computers through the Monitoring Templates. It consists of charts that display detailed statistics for the selected Custom performance metrics, Date Range, and Computers.
You can choose up to 10 different performance counters to display simultaneously for one or more computers. This allows you to select specific performance counters on multiple computers and show them in one chart to easily compare their performance. In addition, you can choose multiple performance counters for one computer and identify problems causing performance issues on that computer.
Here are some tricks that will help you use this report more efficiently:
- Hovering over chart will display a specific tooltip for the selected series. The tooltip displays details about the data point the mouse is currently hovering over, such as the computer name, date and time values, and value of the data point.
- Use the mouse wheel to zoom in and out to view the chart data with maximum precision.
- When displaying data for multiple computers on one chart, click on a specific series to get detailed info about its values (e.g., last, minimum, maximum and average values).
- Use the Date Range and performance counters filters to refine results and quickly navigate to the desired data.
Every chart, i.e. performance counters, in this report has threshold lines with color ranges. This makes them very useful in the initial step to identify performance bottlenecks and for further detailed analysis. When you click on a specific chart series, assuming the data for multiple computers is displayed, the associated warning and critical thresholds will be shown. The specified warning and critical thresholds are now visualized. These data are retrieved from the Monitoring Templates that are applied to computers and can be explored or re-configured on the Administration > Monitoring Templates.
The performance counters filter is a brand-new filter that is only available on the Detailed Analysis report. It consists of three components: a search bar, four status filters, and a server/performance counter/instance tree.
Each of the available components can be selected and shown separately or combined with the other components. By checking a specific performance counter or instance in the tree filter, a chart will appear showing the values for that counter or instance.
The performance counters filter is designed to be a heritable tree structure. Checking the parent node will automatically check and display all the child nodes.
The maximum number of different counters that can be displayed at the same time is 10. If the same performance counter is selected on multiple computers, all the values will be displayed on the same chart.
You can use the search bar to filter computers and performance counter components (e.g., categories, counters, and instances) by their names. Checking or unchecking a specific performance counter status in the status filter will display or hide all the counters and instances that correspond to that status.
The Detailed Analysis report will display custom performance data only if one or more Monitoring Templates has been created and assigned to specific computers or computer groups.
The tree structure status values on the Detailed Analysis report change colors depending on the defined thresholds for the selected performance counters, which are configured through the Monitoring Templates category on the Administration tab.
Custom performance counters can have four different statuses when used to monitor computers:
- Healthy (green): All performance counter values are within the desired limits.
- Warning (yellow): One or more performance counter values has crossed the defined warning threshold.
- Critical (red): One or more performance counter values has crossed the defined critical threshold.
- Missing (gray): No performance counters can be found in the category.
If data collection is still in progress or if no performance counters can be found on monitored computers that have one or more Monitoring Templates assigned, they will be grayed out and shown as Missing.
The Alert Overview report displays all critical and warning alerts generated within the selected Date Range. It will help you get a clear overview of all the alerts raised in your environment.