Jetstress: Disk sub-system thresholds


Performance counter thresholds are updated. But, there are some differences from Exchange 2003 Performance Troubleshooting Guide for some reasons.


Database Disks:               


·         Average Disk sec/Read (or I/O Database Reads Average Latency): The average should be below 20 ms. Spikes (maximum values) should not be higher than 50 ms; more than 6 violations fail the test.


Transactional Log Disks:


·         Average Disk sec/Write (or I/O Log Writes Average Latency): The average should be below 10 ms. Spikes (maximum values) should not be higher than 50 ms; more than 6 violations fail the test.


Here are the latest performance counter thresholds: 










































Disk Sub-system Thresholds


Performance


Stress


Troubleshooting Guide


Data Disk Read Latency (average)


20 ms


20 ms


Data Disk Read Latency (maximum)


50 ms


100 ms


50 ms


Data Disk Write Latency (average)


(n/a)


20 ms


Data Disk Write Latency (maximum)


50 ms


Log Disk Read Latency (average)


(n/a)


5 ms


Log Disk Read Latency (minimum)


50 ms


Log Disk Write Latency (average)


10 ms


10 ms


Log Disk Write Latency (maximum)


50 ms


100 ms


50 ms


 



































Host Computer Thresholds


Performance/Stress


% Processor Time (maximum)


90%


% Processor Time (average)


80%


Available Memory Bytes (minimum)


50 MB


Page Table Entries (minimum)


5000


Transition Pages Repurposed/Sec (average)


100


Memory Pages Per Sec (average)


100


Pool Nonpaged Bytes (maximum)


75 MB


Pool Paged Bytes (maximum)


180 MB


Database Page Fault Stalls Per Sec (average)


1.0


  


Reference: Troubleshooting Exchange Server 2003 Performance


This posting is provided “AS IS” with no warranties, and confers no rights.

Comments (1)

  1. Unified Messaging in Exchange Server 2007 Really Works ESEUTIL for Microsoft Exchange, Part 1 The Basics