perfmon性能计数器是基于相同的东西“在引擎盖下”作为xperf使用的ETW事件？ [英] Are perfmon performance counters based on the same thing "under the hood" as the ETW events used by xperf?

问题描述

最近我已经熟悉了perfmon和xperf。 Perfmon使用性能计数器，xperf使用ETW（窗口的事件跟踪）。 Perfmon有提供数据的对象，而xperf则使用provider组。作为这个领域的新手，我询问是否有人可以告诉我perfmon使用的性能计数器是否真的基于与xperf使用的ETW事件相同的事情，如果是，则可以使连接清楚。如果他们不是基于同样的事情，你可以解释它们是如何不同的？

解决方案

Perf计数器和 ETW 是两个不同的东西，它们没有底层的基础设施。

计数器用于提供
信息，说明
操作系统或应用程序
服务或驱动程序的运行情况。
计数器数据可以帮助确定系统
瓶颈和微调系统和
应用程序性能。操作
系统，网络和设备提供
计数器数据，应用程序可以
消费，为用户提供
图形视图，系统
的执行情况

Windows的事件跟踪（ETW）是一个
高效的内核级跟踪
工具，可让您记录内核或
应用程序定义的事件到日志
文件。

它们都可以用于性能分析，但是ETW提供了一个审计跟踪（如传统的用户模式日志文件），而PerfMon提供了应用程序中的当前统计信息（当前队列长度）或其生命周期中的聚合数据（例如平均吞吐量）的视图， ，发送的总字节数）。

I have recently been becoming acquainted with perfmon and with xperf. Perfmon uses performance counters and xperf uses ETW (event tracing for windows). Perfmon has objects that provide data, whereas xperf uses groups of "providers". As a newbie into this area I am asking whether anyone could tell me whether the performance counters used by perfmon are really based on the same thing under the hood as the ETW events used by xperf and if so, could you make the connection clear. If they aren't based on the same thing, could you explain how they are different?

解决方案

Perf counters and ETW are two different things, they share no underlying infrastructure.

Counters are used to provide information as to how well the operating system or an application, service, or driver is performing. The counter data can help determine system bottlenecks and fine-tune system and application performance. The operating system, network, and devices provide counter data that an application can consume to provide users with a graphical view of how well the system is performing.

Event Tracing for Windows (ETW) is an efficient kernel-level tracing facility that lets you log kernel or application-defined events to a log file.

They can both be used for performance analysis, but ETW provides an audit trail of behaviour inside the app over time (like a traditional user-mode logfile), whereas PerfMon provides a view of either current statistics in the application ('current queue length'), or aggregated data over its lifetime (such as 'average throughput', 'total number of bytes sent').

这篇关于perfmon性能计数器是基于相同的东西“在引擎盖下”作为xperf使用的ETW事件？的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！