To determine which changes or new software features need to be added to increase application
availability, Microsoft has made an effort to gather information about performance in terms of reliability.
To do this, monitoring tools and feedback loops are used to provide an accurate understanding of the
behavior and reliability of Windows in mission-critical environments and to illustrate the challenges that
are faced.
Microsoft collects detailed system information—including the specific causes of downtime—from
hundreds of servers throughout the company’s internal IT infrastructure and thousands of customers
systems. This information is combined with similar data collected from customer environments and is
used to generate a data-driven assessment of the operating system’s performance. Requirements are
determined based on the leading causes of downtime and are fed back to the Windows development
organization for the necessary product improvements.
The following chart shows common causes of server downtime with Windows 2000 Server.
The Windows operating system includes several built-in features designed to help customers measure
server reliability and better understand the causes of downtime:
· Windows Error Reporting is a configurable feature in Windows that transmits information about
operating system and application crashes to Microsoft, where it is used to drive further product
improvement.
· The Shutdown Event Tracker (SET) provides a way to consistently track why a server is shut down
or rebooted. System administrators can choose to have the information forwarded to Microsoft so
that it can be used to further understand customer environments and their causes of downtime.
No comments:
Post a Comment