Cache Memory Usage Test

This test reports how well the objects in the cache have been utilized.

Target of the test : A Hitachi AMS storage device

Agent deploying the test : A remote agent

Outputs of the test : One set of results for every cache on the storage device.

Configurable parameters for the test
Parameter Description

Test Period

How often should the test be executed.

Host

The IP address of the storage device.

UnitName

Specify the name of the storage unit to be monitored. To determine the unit name registered with the Storage Navigator Client for the AMS SAN device, run the following commands, one after another, from the C:\Program Files\Storage Navigator Modular CLI directory on the Storage Navigator Client:

startnsmen

auunitref

The output of the command includes the Name of the storage unit. DF700M_75011118 is a sample unit name.

MCType

Indicate the machine type. The value can be AMS or WMS. By default, this parameter is set to AMS.

SNClientLocation

Specify the full path to the install directory of the storage navigator.

Timeout

Indicate the duration (in minutes) for which this test should wait for a response from the storage device. By default, this is set to 5 minutes.
Measurements made by the test
Measurement Description Measurement Unit Interpretation

Write pending rate

Indicates the percentage of writes to this cache that are currently pending.

Percent

A high value could be a cause for concern, as it could indicate a bottleneck while writing to the cache.  

Clean queue usage rate

Indicates the percentage of the this cache’s clean queue currently utilized.

Percent

Since all pending read/write requests to a cache typically reside in a queue, excessive queue usage (i.e., a high value for this measure) is often an indication that too many requests are yet to be processed by the cache and hence, are still in queue. This could be a result of a processing bottleneck on the cache or a cache overload. You might have to investigate further to diagnose the root-cause of this anomaly.

On the other hand, if the values of these measures are low, it is a sign of good cache health.  

Middle queue usage rate

Indicates the percentage of this cache’s middle queue currently utilized.

Percent

Physical queue usage rate

Indicates the percentage of the cache’s physical queue currently utilized.

Percent

Total queue usage rate

Indicates the percentage of the cache queue currently utilized.

Percent