Using the Performance Monitor data graphs

Plotting the data on graphs

Before you begin

• Required role: Storage Administrator (Performance Management)

Procedure

1. Select Analytics in Explorer, and select Performance Monitor in the tree to open the Performance Monitor window.

2. Click Monitor Performance to open the Monitor Performance window.

3. Specify the data range and time period to be displayed on the graphs:

1. In the Data Range field, select the desired data range.

(VSP G1x00 and VSP F1500) Long-Range or Short-Range. If you want to display real-time data, you must
select Short-Range.

2. In the Time Range field, select Set Range to display data from a specific period of time, or select Use Real
Time to display real-time data as it is being collected.
(VSP G1x00 and VSP F1500) The Use Real Time option can only be specified when Short-Range is
selected.

3. If you selected Set Range, specify the starting date and time and ending date and time for the range of data
in the From and To fields.

4. Specify the monitor data to be displayed on the graphs:

1. In the Performance Objects area select items in the Object field.

Select items in the left field and then select detailed items in the right field. Detailed item changes by the
items selected in the left field.

2. Select items in the Monitor Data field.

Select items in the left field and then select detailed items in the right field.

3. Select the monitor data to be graphed in the Available Objects list.

4. Click Add to add the monitor data to the Selected Objects list.

5. If you need to remove monitor data from the Selected Objects list, select the object and click Remove.

https://knowledge.hitachivantara.com/Documents/Management_Software/SVOS/8.1/Performance_Optimization/Using_the_P…
Updated: Sun, 19 Jan 2020 07:31:20 GMT
Powered by
1
 5. Click Apply to display the selected monitor data on graphs in the Monitor Performance window.

Adding a new graph

Before you begin

• Required role: Storage Administrator (Performance Management)

Procedure

1. Select Analytics in Explorer, and select Performance Monitor in the tree to open the Performance Monitor window.

2. Click Monitor Performance to open the Monitor Performance window.

3. Display the graph panel in the Monitor Performance window.

For details, see Displaying the monitor data on graphs.

4. Click Add Graph to open the Add Graph window.

If a warning message is displayed when you click the Add Graph button, you are already displaying the maximum
number of graphs. To add another graph, you must first delete an existing graph. For details, see Deleting a graph
from the graph panel.

5. Specify the monitor data to be graphed by selecting the objects on the left and adding them to the Selected
Objects list on the right.
For details, see Displaying the monitor data on graphs.

6. Click OK to add the specified graph.

Changing the data displayed on a graph

Before you begin

• Required role: Storage Administrator (Performance Management)

Procedure

1. Select Analytics in Explorer, and select Performance Monitor in the tree to open the Performance Monitor window.

2. Click Monitor Performance to open the Monitor Performance window.

3. Display the graph panel in the Monitor Performance window.

For details, see Displaying the monitor data on graphs.

4. Click Edit Performance Objects to open the Edit Performance Objects window.

5. Edit the monitor data being displayed on the graph by adding objects to and deleting objects from the Selected

https://knowledge.hitachivantara.com/Documents/Management_Software/SVOS/8.1/Performance_Optimization/Using_the_P…
Updated: Sun, 19 Jan 2020 07:31:20 GMT
Powered by
2
 Objects list on the right.
For details, see Displaying the monitor data on graphs.

6. Click OK to display the graph with the newly specified monitor data.

Changing the time period displayed on the graph panel

Before you begin

• Required role: Storage Administrator (Performance Management)

Procedure

1. Select Analytics in Explorer, and select Performance Monitor in the tree to open the Performance Monitor window.

2. Click Monitor Performance to open the Monitor Performance window.

3. Display the graph panel in the Monitor Performance window.

For details, see Displaying the monitor data on graphs.

4. Click Edit Time Range to open the Edit Time Range window.

5. Enter the desired starting date and time in the From field. Enter the desired end date and time in the To field.

6. Click OK to display the graph panel with the newly selected time period.

Deleting a graph from the graph panel

Before you begin

• Required role: Storage Administrator (Performance Management)

Procedure

1. Open the Performance Monitor window.

1. Select Analytics in Explorer.

2. Select Performance Monitor in the tree to open the Performance Monitor window.

2. Click Monitor Performance to open the Monitor Performance window.

3. Display the graph panel in the Monitor Performance window.

For details, see Displaying the monitor data on graphs.

4. Click Delete Graph in the lower right, or click the close icon in the upper right corner.

https://knowledge.hitachivantara.com/Documents/Management_Software/SVOS/8.1/Performance_Optimization/Using_the_P…
Updated: Sun, 19 Jan 2020 07:31:20 GMT
Powered by
3
 5. When the confirmation message appears, click OK to delete the graph.

About the data graphs

The Hitachi Performance Monitor data graphs are displayed on panels with the graph on the left and the legend for the
graph on the right. Hitachi Performance Monitor can display up to four panels with up to four graphs in each panel for a
maximum of 16 graphs at the same time.

• You can change the size of each graph panel by clicking the icon in the upper right of the panel.
• You can plot up to 8 sets of data with one graph panel and up to 16 sets of data with four graph panels.
• You can display the value of each data point by placing the cursor on the data point on the graph.
• You can adjust the unit of scale on the vertical axis using the vertical scale selector on the upper left of the graph.
For example, you might need to increase the vertical scale to see data with high values.

In some cases, you might need to adjust the vertical scale to display the data properly. For example, if the scale is
too small, a single data line might be so thick that the entire graph is painted in a single color.
• You can display or hide data points on the graph panel by clicking the legend on the right of the graph. Each graph
must display at least one set of data, so you cannot hide all of the data.
• When you are viewing data for the MP blade or unit in real time (Time Range is set to Use Real Time), the MP
blade or unit names are displayed as text links. Click the text link to display the resources assigned to an MP blade
or unit of top 20 in usage rates on the detailed window.

Cautions and restrictions for usage statistics

• (VSP G1x00 and VSP F1500) Retention of short-range and long-range usage statistics

Usage statistics for the last six months (186 days) are displayed in long-range monitoring, and usage statistics for
up to the last 15 days are displayed in short-range monitoring. Usage statistics outside of these ranges are deleted
from the SVP. In short range monitoring, results are retained for the last 1 to 15 days depending on the specified
sampling interval. If the retention period has passed since a monitoring result was obtained, the previous result has
been deleted from the SVP and cannot be displayed.
• (VSP Gx00 models and VSP Fx00 models) Retention of usage statistics

Usage statistics for up to the last 15 days are displayed in monitoring. Usage statistics outside of these ranges are
deleted from the SVP.
• Statistics for periods of high I/O workload

If the host I/O workload is high, the storage system gives higher priority to I/O processing than to monitoring. If this
occurs, some monitoring data might be missing. If monitoring data is missing frequently, use the Edit Monitoring
Switch window to lengthen the sampling interval. For details, see Starting monitoring.
• (VSP G1x00 and VSP F1500) Volumes and CU ranges

https://knowledge.hitachivantara.com/Documents/Management_Software/SVOS/8.1/Performance_Optimization/Using_the_P…
Updated: Sun, 19 Jan 2020 07:31:20 GMT
Powered by
4
 The volumes to be monitored by Performance Monitor are specified by control unit (CU). If the range of used CUs
does not match the range of CUs monitored by Performance Monitor, usage statistics might not be collected for
some volumes.
• Reverse resync operations

When you run the CCI horctakeover command, the pairresync-swaps command for a UR pair, or the BCM
YKRESYNC REVERSE command for a URz pair, the primary and secondary volumes are swapped. You can
collect the before-swapped information immediately after you run any of the commands. Invalid monitoring data will
be generated for a short time but will be corrected automatically when the monitoring data gets updated. The invalid
data will temporarily be generated when the volume used for a secondary volume is used as a primary volume after
a UR pair or URz pair is deleted.

• (VSP G1x00 and VSP F1500) When SVP High Availability Feature is installed and the SVP is duplexed, if you
switch the master SVP and the standby SVP, the long-range monitoring data is kept, but the short-range monitoring
data is deleted. If you ask maintenance personnel to switch the master and standby SVPs for microcode upgrade or
other maintenance purposes, execute Export Tool beforehand as necessary, and acquire the short-range
monitoring data.
• Display of monitoring data immediately after monitoring starts or immediately after the sampling interval is
changed

Monitoring data cannot be displayed within the first two sampling intervals after the monitoring starts or the
sampling interval is changed because no monitoring data has accumulated. For instance, if the sampling interval is
set or changed to 15 minutes, monitoring data is not accumulated for up to 29 minutes after this setting is made.
• Display of monitoring data during high SVP workload

If the SVP is overloaded, the system might require more time than the sampling interval allows to update the display
of monitoring data. If this occurs, a portion of monitoring data is not displayed. For example, suppose that the
sampling interval is 1 minute, and the display in the Performance Management window is updated at 9:00 and the
next update occurs at 9:02. In this case, the window (including the graph) does not display the monitoring result for
the period of 9:00 to 9:01. This situation can occur when the following maintenance operations are performed on the
storage system or on the Device Manager - Storage Navigator PC:
◦ Adding, replacing, or removing cache memory.
◦ Adding, replacing, or removing data drives.
◦ Changing the storage system configuration.
◦ Replacing the microcode or firmware.
• Pool-VOLs

Pool-VOLs of Thin Image, Dynamic Provisioning, and Dynamic Provisioning for Mainframe are not monitored.
• Margin of error

The monitoring data might have a margin of error.

https://knowledge.hitachivantara.com/Documents/Management_Software/SVOS/8.1/Performance_Optimization/Using_the_P…
Updated: Sun, 19 Jan 2020 07:31:20 GMT
Powered by
5
Causes of invalid monitoring data
If the value of monitoring data in CSV files is less than 0 (zero), consider the following causes:

Invalid values of
Probable causes
monitoring data

The value (-1) indicates that Performance Monitor failed to obtain monitoring data.
Probable reasons are:

• Performance Monitor attempted to obtain statistics when an operation for restarting

the storage system is in progress.
• Performance Monitor attempted to obtain statistics when a heavy workload is
imposed on the storage system.
• There is no volume in a parity group.
• Just after the CUs to be monitored were added, Export Tool failed to save files that
contain monitoring data for all volumes or journal volumes used by remote copy
software (TC, TCz, UR, URz) and GAD. For details about the files, see Remote
copy operations by TC/TCz and monitoring data by GAD (whole volumes), Remote
copy by UR and URz (whole volumes), and Remote copy by UR and URz (at
journals).
• (VSP G1x00 and VSP F1500) If no CU is specified as a monitoring target, the value
(-1) is displayed as the monitoring data.
• (VSP G1x00 and VSP F1500) If Disable is selected to stop monitoring in the
The monitoring data in
Monitoring Switch field on the Monitoring Options window and longrange is
the CSV file includes (-1). specified as the sampling interval, the monitoring data for the period when
Performance Monitor stops monitoring is (-1).
• (VSP G1x00 and VSP F1500) If you added the CU during monitoring, specified
longrange as the sampling interval, and collected monitoring data, the value (-1) is
displayed as the monitoring data before the CU was added.
• (VSP G1x00 and VSP F1500) If the CU number is not the monitoring target object,
Performance Monitor cannot obtain monitoring data from the CU. However, when
the RemoteCopy, Universal Replicator, or URJNL operand is specified for the group
subcommand, the value (-1) is not displayed as the monitoring data even if the CU
number is not the monitoring target object. In that case, data on the monitored CU
is added up and output to the CSV file.
• (VSP G1x00 and VSP F1500) If all CU numbers are not the monitoring target, the
value (0) is displayed as the monitoring data.
• (VSP G1x00 and VSP F1500) If the mainframe host runs a command to clear
usage information, the monitoring data counter is cleared. As a result, the
monitoring data at that point becomes disabled, and the value (-1) displays as the
monitoring data.

https://knowledge.hitachivantara.com/Documents/Management_Software/SVOS/8.1/Performance_Optimization/Using_the_P…
Updated: Sun, 19 Jan 2020 07:31:20 GMT
Powered by
6
 Invalid values of
Probable causes
monitoring data

The value (-3) indicates that Performance Monitor failed to obtain monitoring data for the
following reason:
The monitoring data in
the CSV file includes (-3). If IOPS is 0 (zero), the Response Time that is included in the monitoring data for LUs,
LDEVs, ports, WWNs, or external volumes is (-3). Because IOPS is 0 (zero), the
average response time becomes invalid.

The value (-4) indicates that Performance Monitor failed to obtain monitoring data for the
following reason:

The monitoring data in

If the period for the monitoring data that is specified with Export Tool does not match the
the CSV file includes (-4).
collecting period for monitoring data, Export Tool cannot collect the monitoring data. If
data of SVP is updated while the monitoring data is being collected, the collected
monitoring data near the collection start time is (-4).

When the CU number is not the monitoring target object, Performance Monitor cannot
obtain monitoring data from the CU.

If the PG, LDEV, LU, RCLU, RCLDEV, URLU, or URLDEV operand is specified, the
value of the monitoring data is "-5". To solve this problem, specify the CU as the
The monitoring data in
monitoring target object by using the Monitoring Options window of Performance Monitor
the CSV file includes "-5".
(not by using Export Tool).

If the RemoteCopy, Universal Replicator, or URJNL operand is specified, the value "-5"
is not output in the monitoring data though the CU number is not the monitoring target
object. In this case, data on monitored CUs are summed up and output into the CSV file.

Next step
Export the performance data as needed for analysis and for retention of long-term data

https://knowledge.hitachivantara.com/Documents/Management_Software/SVOS/8.1/Performance_Optimization/Using_the_P…
Updated: Sun, 19 Jan 2020 07:31:20 GMT
Powered by
7