Windows® Internals, Sixth Edition, Part 1

The only way to specify a starting priority class for a process is with the start command in the Windows command prompt. If you want to have a program start every time with a specific priority, you can define a shortcut to use the start command by beginning the command with cmd /c. This runs the command prompt, executes the command on the command line, and terminates the command prompt. For example, to run Notepad in the low-process priority, the shortcut is cmd /c start /low Notepad.exe.

EXPERIMENT: Examining and Specifying Process and Thread Priorities

Try the following experiment:

From an elevated command prompt, type start /realtime notepad. Notepad should open.

Run Process Explorer, and select Notepad.exe from the list of processes. Double-click on Notepad.exe to show the process properties window, and then click on the Threads tab, as shown here. Notice that the dynamic priority of the thread in Notepad is 24. This matches the real-time value shown in the following image.

Task Manager can show you similar information. Press Ctrl+Shift+Esc to start Task Manager, and click on the Processes tab. Right-click on the Notepad.exe process, and select the Set Priority option. You can see that Notepad’s process priority class is Realtime, as shown in the following dialog box:

Windows System Resource Manager

Windows Server 2008 R2 Enterprise Edition and Windows Server 2008 R2 Datacenter Edition include an optionally installable component called Windows System Resource Manager (WSRM). It permits the administrator to configure policies that specify CPU utilization, affinity settings, and memory limits (both physical and virtual) for processes. In addition, WSRM can generate resource utilization reports that can be used for accounting and verification of service-level agreements with users.

Policies can be applied for specific applications (by matching the name of the image with or without specific command-line arguments), users, or groups. The policies can be scheduled to take effect at certain periods or can be enabled all the time.

After you set a resource-allocation policy to manage specific processes, the WSRM service monitors CPU consumption of managed processes and adjusts process base priorities when those processes do not meet their target CPU allocations.

The physical memory limitation uses the function SetProcessWorkingSetSizeEx to set a hard-working set maximum. The virtual memory limit is implemented by the service checking the private virtual memory consumed by the processes. (See Chapter 10 in Part 2 for an explanation of these memory limits.) If this limit is exceeded, WSRM can be configured to either kill the processes or write an entry to the Event Log. This behavior can be used to detect a process with a memory leak before it consumes all the available committed memory on the system. Note that WSRM memory limits do not apply to Address Windowing Extensions (AWE) memory, large page memory, or kernel memory (nonpaged or paged pool).

Thread States

Before you can comprehend the thread-scheduling algorithms, you need to understand the various execution states that a thread can be in. The thread states are as follows:

Ready. A thread in the ready state is waiting to execute (or ready to be in-swapped after completing a wait). When looking for a thread to execute, the dispatcher considers only the pool of threads in the ready state.

Deferred ready. This state is used for threads that have been selected to run on a specific processor but have not actually started running there. This state exists so that the kernel can minimize the amount of time the per-processor lock on the scheduling database is held.

Standby. A thread in the standby state has been selected to run next on a particular processor. When the correct conditions exist, the dispatcher performs a context switch to this thread. Only one thread can be in the standby state for each processor on the system. Note that a thread can be preempted out of the standby state before it ever executes (if, for example, a higher priority thread becomes runnable before the standby thread begins execution).

Running. Once the dispatcher performs a context switch to a thread, the thread enters the running state and executes. The thread’s execution continues until its quantum ends (and another thread at the same priority is ready to run), it is preempted by a higher priority thread, it terminates, it yields execution, or it voluntarily enters the waiting state.