F5 Disk I/O Test

This test auto-discovers the disks of the F5 traffic manager and reports how well the I/O operations were performed on each disk and in the process, identify delays in I/O operations. In addition, this test also reports the total amount of data transacted when performing read and write operations. This way, this test not only enables administrators to promptly detect current or probable latencies in I/O operations on each disk, but will also help them figure out which I/O activity was most latent – i.e., whether read operations or write operations?

Target of the test : A Big-IP/F5 Local Traffic Manager

Agent deploying the test : A remote agent

Outputs of the test : One set of results for each disk inserted in the slot of the target F5 traffic manager

Configurable parameters for the test
Parameter	Description
Test Period	How often should the test be executed
Host	The IP address of the traffic manager
Port	The port number on which the traffic manager is listening. By default, this is NULL.

Measurements made by the test
Measurement	Description	Measurement Unit	Interpretation
Total I/O Operations	Indicates the total number of read and write operations performed on this disk.	Operations	A high value is desired for this measure. A consistent decrease in this value could indicate a processing bottleneck.
Read operations	Indicates the number of read operations performed on this disk.	Operations	If the value of the Total I/O Operations measure dips consistently, then, you may want to time-correlate that measure with the value of these two measures to know what is causing the steady decline in disk performance – read operations or write operations? This way, you can figure out when the slowdown actually occurred – when reading or when writing?
Write operations	Indicates the number of write operations performed on this disk.	Operations
Merged read operations	Indicates the number of read operations that were merged on this disk.	Operations
Merged write operations	Indicates the number of write operations that were merged on this disk.	Operations
Read data	Indicates the amount of data that has been read from this disk.	MB	Ideally, the value of these measures should be high. A consistent decrease in their value is an indication of a processing bottleneck. When users complaint of delays when accessing the disk, you can compare the value of each of these measures across the disks to know which disk is experiencing the slowdown. You can then compare the value of these measures for that disk to know when the slowdown occurred – when reading from or when writing to the disk?
Write data	Indicates the amount of data that has been written to this disk.	MB
Average read latency	Indicates the average time taken to read the data from this disk.	Milliseconds	Compare the value of this measure across the disks to which disk is catering the read requests at a faster pace.