eG Monitoring
 

Measures reported by LHClusterPerfTest

This test reports how well the cluster handles the I/O requests it receives. In addition, this test also reveals the time taken by the cluster to complete the read and write operations and the number of I/O operations that are pending on the cluster. This way, the test helps administrators to find out the processing bottleneck (if any) and take remedial actions before the slowness in I/O processing affects the performance of the storage system.

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

The measures made by this test are as follows:

Measurement Description Measurement Unit Interpretation
IO_reads Indicates the rate at which read I/O operations were performed on the cluster during the last measurement period. Operations/sec  
IO_writes Indicates the rate at which write I/O operations were performed on the cluster during the last measurement period. Operations/sec  
Total_IO Indicates the total read and write I/O operations performed on the cluster during the last measurement period. Operations/sec This measure serves as a good indicator of the I/O processing ability of the cluster. A consistent drop in this value is hence a cause for concern, as it indicates a processing slowdown.
Data_reads Indicates the rate at which data was read from the cluster. MB/sec Comparing the value of these measures across the clusters will clearly indicate which cluster is the busiest in terms of the rate at which data is read and written - it could also shed light on irregularities in load balancing across the clusters.
Data_writes Indicates the rate at which data was written to the cluster. MB/sec
Data_processed Indicates the overall rate at which data was read from and written to the cluster. MB/sec  
Pending_read_ops Indicates the rate at which read I/O operations were pending on the cluster during the last measurement period. Operations/sec  
Pending_write_ops Indicates the rate at which write operations were pending on the cluster during the last measurement period. Operations/sec  
Pending_IO_operation Indicates the overall rate at which read and write I/O operations were pending on the cluster during the last measurement period. Operations/sec  
Read_latency Indicates the time taken to complete the read I/O operations on the cluster. Millisec  
Write_latency Indicates the time taken to complete the write I/O operations on the cluster. Millisec  
Cache_hits Indicates the rate which read operations were serviced from the cache of the cluster. Operations/sec Ideally, the value of this measure should be high. A consistent drop in cache hits and a steady increase in cache misses during the same time frame is indicative of ineffective read cache usage, which can lead to a slowness in read request servicing.