Specifications

www.ibm.com/support/techdocs 41

Summary of Best Practices for Storage Area Networks

The key question of what needs to be monitored is also the most difficult question to

answer. “It depends.” There are a number of questions which have to be answered

first.

• What monitoring tools are available?

• How much of the SAN environment is appropriate and/or would benefit from

proactive monitoring?

• How often will monitoring be consulted and how will it be used?

Depending on the capability of available tools, the following items are a good starting

point for proactive monitoring.

• Overall Data Rates and I/O Rates

• Backend I/O Rates and Data Rates

• Response Time and Backend Response Time

• Transfer Size and Backend Transfer Size

• Disk to Cache Transfer Rate

• Queue Time

• Overall Cache Hit Rates and Write Cache Delay

• Read-ahead and Dirty Write cache

• Write cache overflow, flush-through and write-through

• Port Data Rates and I/O Rates

• CPU Utilization

• Data Rates, I/O Rates, Response Time, Queue Time for

 Port to Host,

 Port to Disk

• Mirror Rates (while getting synchronized, data change rate, metro or global)

• Peak Read and Write Rates

A key point to understand is that while some or all of the above listed metrics may be

routinely collected by a management application, the administrator must fully

comprehend the meaning and how to interpret a given metric. If this skill is missing,

then the value of performance data gathering is significantly reduced.