eG Monitoring
 

Measures reported by PureVolPerfTest

For users to be able to read from/write data into volumes quickly, adequate space must be available in the volumes and the I/O requests should be processed rapidly by the volumes. Slowdowns in data storage/retrieval can be attributed to storage space contentions experienced by one/more volumes or I/O processing bottlenecks. In the event of such slowdowns, administrators need to swiftly isolate the following:

  • Which volumes are over-utilized?

  • Which volumes are overloaded?

  • Which volumes are experiencing serious latencies?

  • When were these latencies observed most frequently - while reading or writing?

The PureVolPerfTest test provides accurate answers to these questions. With the help of these answers, you can quickly diagnose the root-cause of slowdowns when reading from/writing into a volume.

Outputs of the test : One set of results for each volume of the target FlashArray.

The measures made by this test are as follows:

Measurement Description Measurement Unit Interpretation
readBandwidth Indicates the amount of data utilized for performing read operations on this volume per second. MB/Sec A consistent decrease in the value of these measures for a volume indicates an I/O processing bottleneck.
writeBandwidth Indicates the amount of data utilized for performing write operations on this volume per second. MB/Sec
avgBandwidth Indicates the average amount of data utilized for performing I/O operations on this volume per second. MB/Sec Compare the value of this measure across the volumes to know the volume that uses more bandwidth to process the I/O requests.
readOps Indicates the number of read operations performed on this host per second. IOPS A consistent decrease in the value of these measures for a volume indicates an I/O processing bottleneck.
writeOps Indicates the number of write operations performed on this volume per second. IOPS
avgIOPS Indicates the average number of I/O operations performed on this volume per second. IOPS This measure is a good indicator of I/O load on every host connected to the FlashArray.
readLatency Indicates the time taken by this volume to process the read operations during the last measurement period. Milliseconds Very high values for these measures are indicative of the existence of road-blocks to rapid reading/writing on the volumes.
writeLatency Indicates the time taken by this volume to process the write operations during the last measurement period. Milliseconds
avgLatency Indicates the average time taken by this volume to process the I/O operations during the last measurement period. Milliseconds A consistently high value for this measure is a cause of concern.