EMC Unity Cache Test

This test monitors the usage of the cache supported by the storage system.

Target of the test : An EMC Unity Storage System

Agent deploying the test : A remote agent

Outputs of the test : One set of results for each Storage Processor of the target storage system being monitored.

Configurable parameters for the test
Parameter Description

Test Period

How often should the test be executed.

Host

The IP address of the storage system for which this test is to be configured.

UniSphere CLI Path

The eG agent uses the command-line utility, UEMCli.exe, which is part of the UniSphere Management Suite, to communicate with and monitor the storage device. To enable the eG agent to invoke the CLI, configure the full path to the CLI here.

Username and Password

Provide the credentials of a user possessing 'monitor' role to access the storage system in the Username and Password text boxes.

Confirm Password

Confirm the Password by retyping it in the Confirm Password text box.

Timeout

Specify the duration (in seconds) beyond which this test should time out in the Timeout text box. The default is 300 seconds.

Detailed Diagnosis

To make diagnosis more efficient and accurate, the eG Enterprise embeds an optional detailed diagnostic capability. With this capability, the eG agents can be configured to run detailed, more elaborate tests as and when specific problems are detected. To enable the detailed diagnosis capability of this test for a particular server, choose the On option. To disable the capability, click on the Off option.

The option to selectively enable/disable the detailed diagnosis capability will be available only if the following conditions are fulfilled:

  • The eG manager license should allow the detailed diagnosis capability
  • Both the normal and abnormal frequencies configured for the detailed diagnosis measures should not be 0.
Measurements made by the test
Measurement Description Measurement Unit Interpretation

Clean pages

Indicates the number of cache clean pages in this storage processor.

Number

 

Dirty pages

Indicates the number of dirty cache pages in this storage processor.

Number

These are pages in write cache that have received new data from hosts but have not yet been flushed to disk. While a high value (i.e., a value between 60-80% of the write cache) for this measure is good as it increases the chance of a read coming from cache or additional writes to the same block of data being absorbed by the cache, a very high value - i.e., a value equal to or close to the total number of pages in the write cache - is a sign of bad health, as it indicates that the write cache is over-stressed.

Flushed blocks

Indicates the rate at which data blocks were flushed into the cache of this storage processor.

Blocks/sec

 

Flush rate

Indicates the rate at which flush operations were performed on the cache of this storage processor.

Operations/sec

 

Free pages

Indicates the number of free cache pages in this storage processor.

Number

 

Peak read hits

Indicates the maximum number of read requests serviced by the cache of this storage processor per second.

Hits/sec

The detailed diagnosis of this measure lists the time, name of the storage processor and the read hits during that time period.

Peak read misses

Indicates the maximum number of read requests that failed to be serviced from the cache of this storage processor per second.

Misses/sec

The detailed diagnosis of this measure lists the time, name of the storage processor and the read misses during that time period.

Peak write hits

Indicates the maximum number of write requests to this storage processor that were serviced by the cache per second.

Hits/sec

The detailed diagnosis of this measure lists the time, name of the storage processor and the write hits during that time period.

Peak write misses

Indicates the maximum number of write requests that failed to be serviced to the cache of this storage processor per second.

Misses/sec

The detailed diagnosis of this measure lists the time, name of the storage processor and the write misses during that time period.

Peak dirty cache

Indicates the maximum amount of dirty data available in the dirty cache pages of this storage processor.

MB

 

Average read hits

Indicates the average rate of read requests to this storage processor that were serviced by the cache.

Hits/sec

 

Average read misses

Indicates the average number of read requests that failed to be serviced from the cache of this storage processor per second.

Misses/sec

 

Average write hits

Indicates the average rate of write requests to this storage processor that were serviced by the cache.

Hits/sec

 

Average write misses

Indicates the average number of write requests that failed to be serviced to the cache of this storage processor per second.

Misses/sec

 

Average dirty cache

Indicates the average amount of dirty data available in the dirty cache pages of this storage processor.

MB

 

Free pages percent

Indicates the percentage of free cache pages of this storage processor.

Percentage

The value of this measure is obtained using the formula: (Free Pages/Maximum Pages)*100

A high value is desired for this measure. A sudden/gradual decrease in the value of this measure indicates that the storage processor is running out of free cache pages and hence the cache of the storage processor may reject subsequent entries. This can increase cache misses, increase I/O hits to the storage and subsequently degrade I/O performance of the storage system.

Dirty pages percent

Indicates the percentage of dirty cache pages of this storage processor.

Percentage

A high value for this measure is a cause of concern and is an indication to the administrators to start FAST cache page cleaning.

FAST Cache page cleaning occurs when FAST Cache needs to synchronize the contents of a FAST Cache page with the data’s location in the Pool. This happens when FAST Cache has more recently updated data than the Pool. Cleaning occurs to reduce the amount of FAST Cache dirty pages. When page cleaning starts, a 64 KB FAST Cache page is copied to the Pool in the corresponding location for the data. Once the data is written and the contents of FAST Cache and the Pool are identical, the page in FAST Cache is marked as clean. The data remains in FAST Cache for future use.