Step 5. If you select the Create from a template option, you receive the dialog box shown in Figure 19-14, which enables you to use one of the following templates:
• Basic: Enables you to use performance counters to create a basic data collector set, which you can edit later if necessary.
• System Diagnostics: Enables you to create a report that contains details of local hardware resources, system response times, and local computer
processes. System information and configuration data are also included.
• System Performance: Enables you to create a report that provides details on local hardware resources, system response times, and local computer processes.
• WDAC Diagnostics: Provides trace detailed debug information for
Windows Data Access Components (WDAC) components using BidTrace.
Figure 19-14. Create New Data Collector Set Wizard Enabling You to Use Several Templates
Step 6. Select the desired template and click Browse to locate a template file (XML format) if one exists. Then click Next.
Step 7. Select a location to which you would like the data to be saved (or accept the default location provided), and then click Next.
Step 8. You receive the Create the Data Collector Set? page shown in Figure 19-15. To run the set as a different user, click Change and then select the desired user. To start logging now or configure additional properties, select the option provided. Then click Finish.
Figure 19-15. Create the Data Collector Set? Page Enabling You to Run the Set as Another User or Open the Properties of the Data Collector Set
To create a custom data collector set, use the Create Manually (Advanced) option in Step 4 of the previous procedure and then use the following steps to complete the procedure.
Step 1. After selecting the Create Manually (Advanced) option and clicking Next, you receive the screen shown in Figure 19-16, which enables you to specify the following options:
• Performance Counter: Enables you to select performance objects and
counters to be logged over time. Click Next to specify the performance counters to be logged and the desired sampling interval.
Figure 19-16. Creating Several Types of Logs or Alerts from the Create Manually Option in the Create New Data Collector Set Wizard
• Event Trace Data: Enables you to create trace logs, which are similar to counter logs, but they log data only when a specific activity takes place, whereas counter logs track data continuously for a specified interval.
• System Configuration Information: Enables you to track changes in Registry keys. Click Next to specify the desired keys.
• Performance Counter Alert: Enables you to display an alert when a selected counter exceeds or drops beneath a specified value. Click Next to specify the counters you would like to alert and the limiting value (see Figure 19-17 for an example).
Figure 19-17. Creating an Alert That Informs You When the Processor\%
Processor Time Value Exceeds 75 Percent
Step 2. If you used the Create Data Logs option, select a location to save the log files and then click Next. After clicking Next, or if you used the
Performance Counter Alert option instead, you receive the same dialog box shown previously in Figure 19-15. Make any changes needed and click Finish.
Using Performance Monitor to Create a Data Collector Set
Perhaps the simplest method to create a data collector set is to use a set of counters you have already configured in Performance Monitor. The following steps shows you how:
Step 1. After creating a performance graph as described earlier in this
section, right-click Performance Monitor in the console tree and select New
> Data Collector Set. The Create New Data Collector Set Wizard starts, as previously described.
Step 2. Provide a name for the data collector set and then click Next.
Step 3. Accept the location to which the data is to be saved, or type or browse to the location of your choice, and then click Next.
Step 4. In the Create the Data Collector Set? page, select any required options and then click Finish.
The data collector set is created and placed in the User Defined section. If you select the option to start the data collector set now, logging begins immediately and continues until you right-click the data collector set and choose Stop.
You can view data collected by the data collector set in Performance
Monitor. From the view previously shown in Figure 19-10, select the View Log Data icon (the second icon from the left in the toolbar immediately above the performance graph). In the Source tab of the Performance Monitor Properties dialog box that appears, select the Log Files option and click Add.
Select the desired log file in the Select Log File dialog box that appears, click Open, and then click OK. This displays the selected log in the performance graph.
Optimizing and Troubleshooting Memory Performance
The Memory object includes counters that monitor the computer’s physical and virtual memory. Table 19-2 discusses the most important counters for this object.
Table 19-2. Important Counters for the Memory Object
Counter What It Measures Interpretation and Remedial Tips
Pages/sec
The rate at which data is read to or written from the paging file
A value of 20 or more indicates a shortage of RAM and a possible memory bottleneck.
To view the effect of paging file
performance on the system, watch this counter together with LogicalDisk\% Disk Time. Add RAM to clear the problem.
Available Bytes (KBytes, MBytes)
The amount of physical memory available
A value consistently below 4 MB indicates a shortage of available memory. This might be due to memory leaks in one or more applications. Check your programs for memory leaks. You may need to add more RAM.
Committed Bytes
The amount of virtual memory that has been committed to either physical RAM or running processes
Committed memory is in use and not available to other processes. If the amount of committed bytes exceeds the amount of RAM on the computer, you may need to add RAM.
Pool
Nonpaged Bytes
The amount of RAM in the nonpaged pool system memory (an area holding objects that cannot be
written to disk)
If this value exhibits a steady increase in bytes without a corresponding increase in computer activity, check for an application with a memory leak.
Page Faults/sec
The number of data pages that must be read from or written to the page file per second
A high value indicates a lot of paging activity. Add RAM to alleviate this problem.
In addition to these counters, the Paging File\% Usage counter is of use when troubleshooting memory problems. This counter measures the percentage of the paging file currently in use. If it approaches 100%, you should either increase the size of the paging file or add more RAM.
Lack of adequate memory may also have an impact on the performance of other subsystems in the computer. In particular, a large amount of paging, or reading/writing data from/to the paging file on the hard disk, results in
increased activity in both the processor and disk subsystems. You should monitor counters in these subsystems at the same time if you suspect memory-related performance problems. You learn more about monitoring counters later in the section “Optimizing and Troubleshooting Processor Utilization.”
The paging file is an area on the hard disk that is used as an additional memory location for programs and data that cannot fit into RAM (in other words, virtual memory). By default, the paging file is located at
%systemdrive%\pagefile.sys and has a default initial size of the amount of RAM in the computer plus 300 MB, and a default maximum size of three times the amount of RAM in the computer.
To improve performance on a computer equipped with more than one
physical hard disk, you should locate the paging file on a different hard disk than that occupied by the operating system. You can also increase the size of the paging file or configure multiple paging files on different hard disks. Any of these configurations help to optimize performance by spreading out the activity of reading/writing data from/to the paging files. Note that you should retain a paging file on the system/boot drive to create a memory dump in case of a crash. This memory dump is useful for debugging purposes.
Use the following procedure to modify the configuration of the paging file.
Step 1. In the Search bar or Cortana, type performance and then select Adjust the Appearance and Performance of Windows from the list of options that appears. This opens the Performance Options dialog box.
Step 2. Select the Advanced tab.
Step 3. In the Virtual Memory section of this tab, click Change.
Step 4. As shown in Figure 19-18, the Virtual Memory dialog box displays the disk partitions available on the computer and the size of the paging file on each. To add a paging file to a drive, first clear the Automatically Manage Paging File Size for All Drives dialog box. Select the drive and choose Custom Size to specify an initial and maximum size in MB or System Managed Size to obtain a default size. To remove a paging file, select the drive holding the file and click No Paging File. Note that some programs may not work properly if you choose the No Paging File option. Then click Set.
Figure 19-18. Modifying Paging File Properties from the Virtual Memory Dialog Box
Step 5. Click OK three times to apply your changes and to close the Performance Options and System Properties dialog boxes.
Step 6. Click Restart Now to restart your computer if so prompted.
Optimizing and Troubleshooting Processor Utilization
The processor is the “heart” of the system because it executes all program instructions, whether internal to the operating system or in user-executed applications. The Processor object contains counters that monitor processor performance. Table 19-3 discusses the most important counters for this object.
Table 19-3. Important Counters for the Processor Object
Counter What It
Measures Interpretation and Remedial Tips
% Processor Time
The percentage of time the processor is executing meaningful actions
(excludes the Idle process)
If this value is consistently greater than 85%, the processor could be causing a bottleneck.
You should check the memory counters discussed previously; if these are high,
consider adding more RAM. Otherwise, you should consider adding a faster processor (or an
additional one if supported by your motherboard).
The rate of service requests
from I/O A significant increase in the number of
Interrupts/sec devices that interrupt other processor activities
interrupts, without a corresponding increase in system activity, may indicate some type of hardware failure. Brief spikes are acceptable.
You should also look at the System\Processor Queue Length counter. If the value of this counter exceeds 2, a processor bottleneck may exist, with several programs contending for the processor’s time.
Almost every computer today has more than a single processor “core,” and these will show up as several processors in the performance counters list.
Note that you can use the _Total counter, which provides statistics of all processors combined, or you can select individual processors, which are listed numerically as processor 0, 1, 2, and so on. Including the _Total counter and the individual processors can help you determine if any older, single-threaded applications or processes are causing an issue. Many older applications still in use today do not take advantage of multiple processors and can use only a single processor or core for all their work. If you notice that a single processor is at high utilization, while others are low or idle, you likely have such an application, and it is very busy.
As mentioned in Table 19-3, memory shortages may frequently manifest themselves in high processor activity. It is usually much cheaper and easier to add RAM to a computer than to add a faster or additional processor.
Consequently, you may want to consider this step first when you are experiencing frequent high processor activity.
Optimizing and Troubleshooting Disk Performance
Disk performance is measured by two processor objects: The PhysicalDisk counters measure the overall performance of a single physical hard disk rather than individual partitions. LogicalDisk counters measure the
performance of a single partition or volume on a disk. These counters include the performance of spanned, striped, or RAID-5 volumes that cross physical disks.
PhysicalDisk counters are best suited for hardware troubleshooting. Table 19- 4 describes the most important counters for this object.
Table 19-4. Important Counters for the PhysicalDisk Object
Counter What It Measures Interpretation and Remedial Tips
% Disk Time
The percentage of time that the disk was busy reading or writing to any
partition
A value of over 50% suggests a disk bottleneck. Consider upgrading to a faster disk or controller. Also check the memory counters to see whether more RAM is needed.
Avg. Disk Queue Length
The average
number of disk read and write requests waiting to be performed
If this value is greater than 2, follow the same suggestions as for % Disk Time.
Average Disk
Sec/Transfer
The length of time a disk takes to fulfill requests.
A value greater than 0.3 may indicate that the disk controller is retrying the disk continually because of write failures.
LogicalDisk counters are best suited for investigating the read/write
performance of a single partition. Table 19-5 describes the most important counters for this object.
Table 19-5. Important Counters for the LogicalDisk Object
Counter What It
Measures Interpretation and Remedial Tips
% Disk Time
The
percentage of time that the disk is busy servicing disk requests
A value greater than 90% may indicate a performance problem except when using a RAID device. Compare to Processor\%
Processor Time to determine whether disk requests are using too much processor time.
Average Disk Bytes/Transfer
The amount of data
transferred in each I/O operation
Low values (below about 20 KB) indicate that an application may be accessing a disk
inefficiently. Watch this counter as you close applications to locate an offending
application.
Current Disk Queue Length
The amount of data waiting to be transferred to the disk
A value greater than 2 indicates a possible disk bottleneck, with processes being delayed because of slow disk speed. Consider adding another faster disk.
Disk
Transfers/sec
The rate at which read or write
operations are performed by the disk
A value greater than 50 may indicate a disk bottleneck. Consider adding another faster disk.
% Free Space
Percentage of unused disk space
A value less than about 15% indicates that insufficient disk space is available. Consider moving files, repartitioning the disk, or adding another disk.
Tip
You should log disk activity to a different disk or computer. The act of
recording performance logs places an extra “hit” on performance for the disk on which logs are recorded. To obtain accurate disk monitoring results,
record this data to a different disk or computer.
Monitoring System Resources
You have learned about several tools in this chapter for monitoring resources and performance of your Windows computers. If you are using an MDM tool such as Microsoft Intune, you can also configure alerts to monitor computers for resource issues. You learned about using Intune alerts in Chapter 13,
“Microsoft Intune.” Figure 19-19 shows some of the many alert types available in Intune for monitoring resources. Alert types are available for monitoring memory, disk, CPU utilization, and others.
Figure 19-19. Some Alert Types Available for Monitoring Computers Using Microsoft Intune
You can perform several tasks associated with performance monitoring and optimization from the command line. The following are several available tools:
• Logman: Manages data collector logs. You can start, stop, and schedule the collection of performance and trace data.
• Relog: Creates new performance logs from data in existing logs by
modifying the sampling rate and/or converting the file format.
• Typeperf: Displays performance data to the command prompt window or to a log file.
You can also use the Perfmon command to start the Performance Monitor from a command line. For information on running these tools, type the command name followed by /? at a command prompt.
Monitoring and Managing Printers
We have already shown you how to share printers and assign permissions to them in Chapter 14, “Configuring File and Folder Access.” This chapter also discusses location-aware printing, which enables you to send documents to a conveniently located printer by adjusting whether to allow Windows to manage your default printer. You also learned about print sharing,
configuring printer access security, and manually managing default printers.
Windows also includes the Printer Management tool in Windows 10. If you have enabled Administrative tools in the Start menu, you can click the Printer Management icon in the folder to open Print Management, which displays the console shown in Figure 19-20. You can also open the Print Management console from the Search bar or Cortana by searching for print, and then selecting Print Management from the list.
Figure 19-20. Monitor and Manage Your Printers Using the Printer Management Console
From the Print Management console, you can view all the printers and printer drivers. You can also add all the print servers available on your network, view the drivers and printers connected to those, and manage the printer queues and jobs. In Figure 19-20, you can see under Custom Filters that the console has provided a way to view at a glance how the printers are doing. In the figure, there are 14 printers, 12 drivers, no printers that are down (not ready), and no printers with jobs. So it’s easy to see what is going on with printing in your environment.
A print server is available by default on any Windows 10 computer or Windows Server. So when you see Print Servers in your Print Management console, it is really referring to any computer that does not have printing disabled. To add a print server, right-click the Print Servers node and select the Add/Remote Servers menu option. From the dialog box, you can browse for a computer on your network or type the computer name into the box.
Click Add to List to add the computer to the list. When you click OK, the computer will show up under the Print Servers node. Refer to Figure 19-20, which has three print servers connected.
You can manage any printer from this interface, as long as you have permissions. You can select one of the folders in Custom Filters or the
Printers node under one of the Print Servers to see the printers. You can right- click one of the folders to view the context menu. This provides many of the same options available from the Control Panel Printers and Scanners applet for managing printers, such as viewing the printer queue, setting sharing, and accessing the device properties pages. If you had administration permissions on a Windows Active Directory, you also have the option to Deploy with Group Policy. As shown in Figure 19-21, this form provides an automatic way to deploy the printer connection to an existing Group Policy Object.
Figure 19-21. Deploying Printer Connections Using Group Policy Objects from the Printer Management Console