Topic Last Modified: 2006-09-06
The Microsoft® Exchange Server Analyzer Tool includes a performance data collection engine that is used to query performance counter objects on computers that are running Exchange 2000 Server or Exchange Server 2003. The performance data collection engine collects data from the Avg Disk sec/Read and Avg Disk sec/Write counters for the PAGEDISK instance of the LogicalDisk performance object to analyze performance data.
If the Exchange Server Analyzer determines that the Avg Disk sec/Read and Avg Disk sec/Write counters exceeded 10 milliseconds during the sample time slice, the Exchange Server Analyzer displays an error.
The LogicalDisk performance object consists of counters such as Avg Disk sec/Read and Avg Disk sec/Write that monitor logical partitions of hard disks. The Avg Disk sec/Read performance counter shows the average time, in seconds, of a read of data from the disk. The Avg Disk sec/Write performance counter shows the average time, in seconds, of a write of data to the disk The average value of the Avg Disk sec/Read and Avg Disk sec/Write performance counters should be under 10 milliseconds.
The PAGEDISK instance of the LogicalDisk performance object on a server represents the page file (Pagefile.sys). Pagefile.sys is a hidden system file on the hard disk that is used by Microsoft Windows® to store temporary data when your system is running low on RAM. The page file is also known as paging file or swap file.
This error indicates that there is a latency problem with the disk hosting the page file.
To resolve this error, take the following steps:
- Identify the cause of the latency. If this issue is caused by
excessive paging it may be caused by server memory issues.
- Consider locating the page file on a dedicated partition or
high speed disk.
- Consider creating a secondary page file on an alternative drive
as described in Microsoft Knowledge Base article 197379,
"Configuring page files for optimization and recovery in Windows
Server 2003, in Windows 2000, and in Windows NT" (http://go.microsoft.com/fwlink/?linkid=3052&kbid=197379).
- Run server maintenance tasks during non-peak times. If you run
maintenance tools, such as Exchange Server Database Utilities
(Eseutil.exe), or tasks, such as mailbox management, during peak
times, you can use memory that would otherwise be used by Exchange.
As a best practice, run these tools and tasks during non-peak times
or during low usage periods.
For More Information
- For more information about disk bottlenecks, see Disk Bottleneck
Detected
- For more information about optimizing page file performance,
see Microsoft Knowledge Base article 197379, "Configuring page
files for optimization and recovery in Windows Server 2003, in
Windows 2000, and in Windows NT" (http://go.microsoft.com/fwlink/?linkid=3052&kbid=197379).
- For more information about Exchange Server performance, see the
Performance and Scalability Guide for Exchange Server 2003
(http://go.microsoft.com/fwlink/?LinkId=47576).
- For more information about troubleshooting Exchange Server
performance issues, see Troubleshooting Microsoft Exchange
Server Performance (http://go.microsoft.com/fwlink/?LinkId=47588).
- For more information about best practices for designing storage
architectures, see Best Practices Common to Multiple
Architectures (http://go.microsoft.com/fwlink/?LinkId=72988).
- For more information about disk sizing, latency, and I/O rates,
see the following blogs on the Microsoft Exchange Team Blog Web
site:
- A few basic concepts in disk sizing (http://go.microsoft.com/fwlink/?linkid=39538)
- Some more thoughts on disk IO and calculations...
(http://go.microsoft.com/fwlink/?linkid=39539)
- How to Calculate Your Disk I/O Requirements (http://go.microsoft.com/fwlink/?linkid=69747)
- A few basic concepts in disk sizing (http://go.microsoft.com/fwlink/?linkid=39538)