NEC Storage Performance Monitor/Optimizer User Manual page 15

data transfer during the unit measurement period and dividing it by the time interval of the measurement period.
(x) Internal Average Transfer Length
This is the average input/output data length of the input/output process performed for the internal disk devices by the
control device of the disk array. The unit is kilobyte. This is calculated by measuring the total of the amount of
internal data transfer and internal I/O count during the unit measurement period and dividing the former by the latter.
(xi) Internal Average Response Time
This is the average response time for the input/output requests issued to the internal disk devices by the control device
of the disk array. The unit is millisecond. This is calculated by measuring the total of the internal response time
and internal I/O count during the unit measurement period and dividing the former by the latter.
(2) Unit to sum up
iSM separately displays values of the monitored items collected in the following units.
(i) Disk array
Statistic information of the entire disk array to be monitored is collected and load status for each disk array is reported.
For this component type, the external I/O Density, external Transfer Rate, external Average Transfer Length, external
Average Response Time, Read Hit Ratio, and Write Hit Ratio can be reported as monitored items.
(ii) Port
For the monitored disk arrays, statistic information related to each host port is collected separately, and load status for
each host port is reported. For this component type, the external I/O Density, external Transfer Rate, external
Average Transfer Length, and external Average Response Time can be reported as monitored items.
In principle, the sum of the values of a monitored item over all the host ports in one disk array is equal to the sum of
the values of the same monitored item over all the disk arrays. However, in a real-world situation, some error may
exist due to varied sampling time of the statistic information which is the basis of the values of the monitored item.
(iii) Logical disk
For the monitored disk arrays, statistic information related to each logical disk is collected separately, and load status
for each logical disk is reported. For this component type, the external I/O Density, external Transfer Rate, external
Average Transfer Length, external Average Response Time, Read Hit Ratio and Write Hit Ratio can be reported as
monitored items.
In principle, the sum of the values of a monitored item over all the logical disks in one disk array is equal to the sum
of the values of the same monitored item over all the disk arrays. However, in a real-world situation, some error may
exist due to varied sampling time of the statistic information which is the basis of the values of the monitored item.
(iv) Physical disk
Disk devices installed in a disk array are called physical disks. Statistic information related to each physical disk
installed in the monitored disk arrays is collected separately, and load status for each physical disk is reported. For
this component type, the internal I/O Density, internal Transfer Rate, internal Average Transfer Length, internal
Average Response Time, and Busy Ratio can be reported as monitored items.
You can set the update frequency (Display Refresh Rate) of display on the Numeric Table and the Time-series Graph by
using the environment setting at system installation.
You can change the setting from the iSM server menu while the iSM server is operating if the performance monitoring
Chapter 1 Overview of Performance Monitoring Function
I-9