EVA LUN Traffic Test

LUN is a Logical Unit Number. A LUN represents a logical abstraction or, if you prefer, virtualization layer between the physical disk device/volume and the applications. They are also referred to as virtual disks. A virtual disk can also be a Snapshot, Snapclone, or replication volume.

This test auto-discovers the LUNs on an EVA storage array, tracks the I/O requests to each LUN, and reveals how well the cache services these requests.

Target of the test : An EVA Storage Array

Agent deploying the test : A remote agent

Outputs of the test : One set of results for each virtual disk on an EVA.

Configurable parameters for the test
Parameter Description

Test Period

How often should the test be executed.

Host

The host for which the test is to be configured.

Port

Refers to the port used by the EVA SAN device.  By default, it is null.

EvaPerfLocation

Specify the full path of the directory in which the evaperf.exe has been installed.

ArrayName

Specify the name of the EVA SAN array that has been configured for monitoring.

SSSULocation

This test uses the EVAPerf software component to auto-discover the device ports on an EVA storage array. The EVA LUN Traffic test on the other hand uses the SSSU console to discover device ports. During auto-discovery, the EVAPerf utility reports the WWN names of the device ports. In contrast, the SSSU console reports the friendly names of the same device ports.

To ensure that device port references are uniform and consistent across tests, you can optionally configure this test to use the SSSU console (instead of EVAPerf) to perform device port discovery. This way, the names of device ports will be the same for this test and the EVA LUN Traffic test. To enable device port discovery via the SSSU console, configure the full path to the location of the SSSU.exe in the SSSULocation text box. If you prefer not to use SSSU.exe for discovering device ports and would rather use the EVAPerf component instead, then, set this parameter to none.

Username,
Password and
Confirm Password

To use the SSSU.exe for discovering the friendly names of device ports, specify the user credentials using which the HP EVA array is to be accessed in the Username and Password text boxes. Confirm the password by retyping it in the Confirm Password text box. If you do not wish to use the SSSU.exe, then set these parameters to none

Timeout

Indicate the duration (in seconds) for which this test should wait for a response from the storage device. By default, this is set to 60 seconds.

Detailed Diagnosis

To make diagnosis more efficient and accurate, the eG Enterprise suite 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

Read requests served by cache

Indicates the rate at which requests to the LUN  were serviced by cache.

Reqs/Sec

The detailed diagnosis of this measure reveals the LUN group to which this LUN belongs.

Data read from cache

Indicates the rate at which this LUN read data from the cache.

MB/sec

 

Latency for read hits

Indicates the time taken by this LUN for reading from the cache.

Ms

A high value could indicate an I/O bottleneck.

Read miss requests

Indicates the rate at which read requests to the LUN were serviced by the disk and not the cache.

Reqs/sec

Ideally, this rate should be low. A high rate indicates a large number of direct disk accesses, which in turn may expose the physical disk to overuse, and may increase the processing overheads.

Read miss data

Indicates the rate at which data was read from the physical disk and not the cache.

MB/sec

Latency for read misses

Indicates the time taken by this test for reading from the physical disk.

Ms

 

Write request rate

Indicates the rate at which write requests were serviced by vDisk.

Reqs/sec

 

Data written to disk

Indicates the rate at which data was written to the disk.

MB/sec

 

Write latency

Indicates the time taken for writing to the disk.

Ms

A high value could indicate an I/O bottleneck.