Skip to main content
NetApp Knowledge Base

Could a storage bottleneck be causing Microsoft Exchange server to be slow?

  1. Last updated
  2. Save as PDF
Views:
1,473
Visibility:
Public
Votes:
1
Category:
ontap-9
Specialty:
perf
Last Updated:

Applies to

  • Microsoft Exchange Server
  • ONTAP 9
  • Data ONTAP 8 7-mode

Answer

  • Typically, to determine if the performance issue is storage related, we need to determine if the Exchange host is seeing any latency from the Logical Unit Number (LUNs) that are mounted
  • From the Exchange host use the Perfmon tool that is included in Windows.
  • Unless Perfmon is showing latency on those virtual disks, the performance issue is typically on the host and not the filer

Additional Information

  • Please start a Perfmon session on the Exchange host and add the following counters:
    • PhysicalDisk\\Average Disk sec/Read
    • PhysicalDisk\\Average Disk sec/Write
    • PhysicalDisk\\Average Disk Queue Length
  • Microsoft has defined the following guidelines for the functions that those disks serve as values for the counters:
Database
PhysicalDisk\\Average Disk sec/Read These values should average less than 20ms and spikes should be less than 50ms
PhysicalDisk\\Average Disk sec/Write These values should average less than 20ms and spikes should be less than 50ms
PhysicalDisk\\Average Disk Queue Length Please note this value
Transaction Logs
PhysicalDisk\\Average Disk sec/Read These values should average less than 5ms and spikes should be less than 50ms
PhysicalDisk\\Average Disk sec/Write These values should average less than 10ms and spikes should be less than 50ms
PhysicalDisk\\Average Disk Queue Length Please note this value
SMTP queues
PhysicalDisk\\Average Disk sec/Read These values should average less than 10ms and spikes should be less than 50ms
PhysicalDisk\\Average Disk sec/Write These values should average less than 10ms and spikes should be less than 50ms
PhysicalDisk\\Average Disk Queue Length Please note this value
Page File Disk
PhysicalDisk\\Average Disk sec/Read These values should average less than 10ms and spikes should be less than 50ms
PhysicalDisk\\Average Disk sec/Write These values should average less than 10ms and spikes should be less than 50ms
PhysicalDisk\\Average Disk Queue Length Please note this value
  • Note: The page file disk should on local storage if it is not currently located there
  • The following counters are not storage specific, but will provide information about how the host is processing Microsoft Exchange requests and help identify an Exchange issue:
    • Memory\\Available Mbytes (MB)
    • Memory\\Pages/sec
    • Processor\\% Processor Time (_Total)
    • System\\Processor Queue Length
    • Network Interface\\Bytes Total/sec
    • Network Interface\\Packets Outbound Errors
    • SMTP Server\\Categorizer Queue Length
    • MSExchangeDSAccess Process\\LDAP Read Time (for all processes)
    • MSExchangeDSAccess Process\\LDAP Search Time (for all processes)
  • From the filer, use qos statistics volume latency show to check the performance compared to the Perfmon

 

NetApp provides no representations or warranties regarding the accuracy or reliability or serviceability of any information or recommendations provided in this publication or with respect to any results that may be obtained by the use of the information or observance of any recommendations provided herein. The information in this document is distributed AS IS and the use of this information or the implementation of any recommendations or techniques herein is a customer's responsibility and depends on the customer's ability to evaluate and integrate them into the customer's operational environment. This document and the information contained herein may be used solely in connection with the NetApp products discussed in this document.