eG Monitoring
 

Measures reported by NetCluLunPerfTest

This test auto-discovers the LUNs configured on the NetApp Cluster, monitors the processing ability of each LUN, and reports the following:

  • Is I/O load uniformly balanced across all LUNs, or is any LUN overloaded?
  • Are the LUNs able to process the I/O requests quickly? Is any LUN experiencing processing bottlenecks?
  • How many errors are encountered by each LUN?

The measures made by this test are as follows:

Measurement Description Measurement Unit Interpretation
Other_ops Indicates the rate at which operations other than read and write were performed on this LUN. Ops/Sec  
Read_ops Indicates the rate at which the read operations were performed on this LUN. Ops/Sec A high value is desired for this measure. A consistent decrease in this value could indicate a processing bottleneck.
Write_ops Indicates the rate at which the write operations were performed to this LUN. Ops/Sec A high value is desired for this measure. A consistent decrease in this value could indicate a processing bottleneck.
Avg_latency Indicates the average time taken for executing an operation in this LUN. Milliseconds A high value indicates that the LUN is taking too long to process the I/O requests to it.

Compare the value of this measure across LUNs to isolate the slow LUNs.
Avg_read_latency Indicates the average time taken for executing an operation in this LUN. Milliseconds A low value is desired for this measure. A high value indicates that the requests take too long to execute which directly affects the performance of the LUNs.
Avg_write_latency Indicates the average time taken to execute a write request in this LUN. Milliseconds  
Scsi_errors Indicates the total number of SCSI errors encountered on this LUN. Number Ideally, the value of this measure should be zero.