Administrator Guide

basic assumption for the round-robin policy is that the data paths are equal. With mixed host support, the data paths may have
different bandwidths or different data transfer speeds.
Least queue depth with subset — The least queue depth with subset policy is also known as the least I/Os or least requests policy.
This policy routes the next I/O request to a data path that has the least outstanding I/O requests queued. For this policy, an I/O
request is simply a command in the queue. The type of command or the number of blocks that are associated with the command are
not considered. The least queue depth with subset policy treats large block requests and small block requests equally. The data path
selected is one of the paths in the path group of the RAID controller module that owns the virtual disk.
Least path weight with subset (Windows operating systems only) — The least queue depth with subset policy is also known as the
least I/Os or least requests policy. This policy routes the next I/O request to a data path that has the least outstanding I/O requests
queued. For this policy, an I/O request is simply a command in the queue. The type of command or the number of blocks that are
associated with the command are not considered. The least queue depth with subset policy treats large block requests and small block
requests equally. The data path selected is one of the paths in the path group of the RAID controller module that owns the virtual disk.
Monitoring system performance
Performance Monitor allows you to track a storage array’s key performance data and identify performance bottlenecks in your system.
You can use Performance Monitor to perform these tasks:
View in real time the values of the data collected for a monitored device. This capability helps you to determine if the device is
experiencing any problems.
Identify when a problem started or what caused a problem by seeing a historical view of a monitored device.
Specify the performance metric and the objects that you want to monitor.
View data in tabular format (actual values of the collected metrics) or graphical format (as line graphs), or export the data to a file.
Three types of performance monitoring exist:
Real-time graphical—Plots performance data on a graph in near real time.
Real-time textual—Shows performance data in a table in near real time.
Background (historical)—Plots graphical performance data over a longer period. You can view background performance data for a
session that is currently in progress or for a session that you previously saved.
This table shows some specific characteristics of each type of performance monitoring:
Table 3. Characteristics of different types of performance monitoring
Type of
Performance
Monitoring
Sampling Interval Length of Time
Displayed
Maximum Number
of Objects
Displayed
Ability to Save
Data
How Monitoring
Starts and Stops
Real-time graphical 5 sec 5 min rolling window 5 No Starts automatically
when AMW opens.
Stops automatically
when AMW closes.
Real-time textual 5-3600 sec Most current value No limit Yes Starts and stops
manually. Also stops
when View Real-
time Textual
Performance
Monitor dialog
closes or AMW
closes.
Background 10 min 7 day rolling window 5 Yes Starts and stops
manually. Also stops
when EMW closes
or firmware
download starts.
Keep these guidelines in mind when using Performance Monitor:
Each time the sampling interval elapses, the Performance Monitor queries the storage array again and updates the data. The impact to
storage array performance is minimal.
26
About your MD Series storage array