% Processor Time (Processor _Total)无法超过100%，但%Processor Time (Process _Total)能超过。
Examining Processor Time Data:
On
multiprocessor systems, the Processor\% Processor Time value reported
by System Monitor will never exceed 100 percent for any particular
processor or thread. On the other hand, the value of the % Processor
Time reported for the Process object can report values over 100
percent; if such values occur, this could indicate that threads of the
process are cumulatively using more than 100 percent of a processor. To
get more detailed information, use the Thread object counters to
analyze the processor time each thread within a process is using.
Investigating other data described in this section, such as DPC
activity or context switching, might help you to interpret high
processor-time values.
多处理器系统中，由系统监视报告的处理器Processor Time%值对任何个别处理器或线程不会超出100%。另一方面，进程对象Processor Time%报告值能超出100%，如果超出就意味着该进程的这些线程在累加地超出100%使用一个处理器。要得到更详尽的信息，使用线程对象计数器去分析该处理器在一个进程里使用每个线程的时间。调查在这部分描述的其它数据，如DPC活动或上下文切换，可能会帮助你去解释高Processor Time%值。
% Processor Time(Processor_Total)
指处理器用来执行非闲置线程时间
的百分比。计算方法是:度量处理器用来执行空闲线程的时间，然后用100%减去该值。(每个处理器有一个空闲线程，该线程在没有其他线程可以运行时消耗周
期)。此计数器是处理器活动的主要指示器，显示在采样间隔期间所观察的繁忙时间平均百分比。应注意，对处理器是否空闲的计算是在系统时钟的内部采样间隔期
间(10ms)执行的。考虑到现在的处理器速度非常快，因此，在处理器可能会用大量时间为系统时钟采样间隔之间的线程提供服务时，%
Processor Time
会低估处理器利用率。当恰好进行采样后即向计时器发出信号时，更可能对应用程序做出不准确地度量，基于工作负荷的计时器应用程序是一个这样的示例。
%Processor Time (Process _Total)
D0是所有进程线程使用处理器执行指令所花的时间百分比。指令是计算机执行的基础单位。线程是执行指令的对象，进程是程序运行时创建的对象。此计数包括处理某些硬件间隔和陷阱条件所执行的代码。
多处理器系统中，对任何处理器或线程，系统监视器报告的Processor\% Processor
Time值不会超出100%。另一方面，进程对象(也就是Process)Processor
Time%值能超出100%；如果超出就意味着该进程的线程累计使用超出了一个处理器的100%。要得到更详尽的信息，使用线程对象计数器去分析在该进程
中每个线程的Processor
Time%值。调查在这部分描述的其它数据，如延迟过程调用(DPC)活动或上下文切换，可能会帮助你去解释高Processor Time%值。
“\Process(…)\%
Processor Time” can go up to N*100 (where N is the number of CPUs)
because it adds up the CPU usage of the requested process across all
the CPUs.
l,T8G&X9D$L$M$N!D2K02.“\Process(_Total)\%
Processor Time” should always be around N*100 (where N is the number
CPUs) because it adds up the CPU usage of each process, including the
idle process.
3.“\Processor(…)\% Processor Time” can go up to 100 because it’s the CPU usage of the requested CPU.
“\Processor(_Total)\% Processor Time” can go up to 100 because it’s the average CPU usage across all CPUs.
Processor(_Total)\% Processor Time
This counter gives the total processor utilization figure for
the server as a whole. For example, on a dual processor system with one
processor pegged out at 100% utilization and the other processor
standing idle, a figure of about 50% would be recorded.
This server metric is useful for gauging overall how busy a
server is or, put in another way, how much spare capacity there is
available (by subtracting from 100%).
If per-processor information is required to be monitored by
Resource Manager, then additional server metrics may be configured for
monitoring:
Processor(0)\% Processor Time
Processor(1)\% Processor Time
... etc.
(See the Resource Manager Administrator’s Guide and the help
within the Presentation Server Console for information on Customizing
Server Metrics).
Process Processor Utilization
In addition to monitoring server metrics Resource Manager also
monitors process activity; one aspect of process activity monitored is
the degree to which the processor(s) is being used.
Processor usage of each process on the server is sampled at
regular intervals using a standard operating system call - note that
the operating system call used returns the sum of processor
time used by all threads within the process. Resource Manager expresses
the processor time used during the sampled interval as a percentage by
dividing this time by the elapsed time since the previous sample.
On a multi-processor system it is quite possible for the
percentage reported to exceed 100% in the case of a multi-threaded
process. For example, on a dual-processor system, a multi-threaded
process in which two threads are performing lengthy processor intensive
calculations would cause a value of about 200% to be recorded.
This process-specific processor utilization is useful for
gauging the state (and possibly health) of the process; if a process
which normally uses little processor time is seen to be consuming a
figure approaching 100%, a likely explanation is that it has
encountered a bug which has caused it to enter an uncontrolled looping
state.