eG Monitoring
 

Measures reported by LHStrSystemPerfTest

This test reveals the following critical statistics for the cluster:

  • CPU and memory utilization;

  • the rate at which the read and write operations were performed on the cluster;

  • the rate at which the I/O operations were in pending state;

  • the time taken by the cluster to complete I/O operations.

Using these metrics, administrators can find out abnormalities, if any, on the cluster at an early stage and prevent the processing bottlenecks.

Outputs of the test : One set of results for the target storage cluster monitored.

The measures made by this test are as follows:

Measurement Description Measurement Unit Interpretation
CPU_busy Indicates the percentage of CPU utilized by the storage cluster. Percent  
CPU_idle Indicates the percentage of CPU that is idle on the storage cluster. Percent  
Memory_util Indicates the percentage of memory utilized for the storage cluster. Percent A low value is preferred for this measure. A high value of this measure indicates a potential space crunch on the cluster.
IO_reads Indicates the rate at which read I/O operations were performed on the storage cluster during the last measurement period. Operations/sec  
IO_writes Indicates the rate at which write I/O operations were performed on the storage cluster during the last measurement period. Operations/sec  
Total_IO Indicates the total read and write I/O operations that were performed on the storage cluster during the last measurement period. Operations/sec  
Data_reads Indicates the rate at which data was read from the storage cluster. MB/sec  
Data_writes Indicates the rate at which data was written to the storage cluster. MB/sec  
Data_processed Indicates the rate at which data was read from and written to the storage cluster. MB/sec  
Pending_IO_operation Indicates the rate at which the I/O operations were moved to pending state during the last measurement period. Operations/sec A low value of preferred for this measure. A high value indicates the processing bottle neck on the cluster.
Read_latency Indicates the time taken to complete read operations on the storage cluster. Millisec The values of these measures should be very low. High values indicate that the cluster is taking much time to complete the I/O operations, which may degrade the performance of the cluster.
Write_latency Indicates the time taken to complete write operations on the storage cluster. Millisec