hIOmon™
I/O Trender Display
Help Information

Brief help information for the following topics related to the hIOmon I/O Trender Display is provided below:

Background Information

The hIOmon "I/O Trender Display" can be used to display - within a graphical format - metrics that have been retrieved by the hIOmon Disk I/O Ranger Display.

These metrics include several of the hIOmon "Performance Threshold Range Metrics", which are a selected set of summarized I/O operation performance metrics that are specifically focused upon the ranges of I/O operations per second (IOPS), megabytes-per-second (MB/s) data transfer rates, response time (RT), I/O operation queue depths/lengths (QD), and data transfer sizes/lengths (DTS) actually observed by the hIOmon I/O Monitor for individual files, devices, and processes.  These metrics are included within the automatically-summarized I/O operation performance information that can be collected by the hIOmon software.

Please note that the hIOmon "Performance Threshold Range Metrics” reflect the actual I/O operation activity observed by the hIOmon I/O Monitor software component.  For instance, the "IOPS 1 - 99" Range Count for read I/O operations reflects the number of one-second intervals during which the hIOmon I/O Monitor actually observed a total number of monitored read I/O operations that was between one and 99 read I/O operations inclusive.

Similarly, the "MBS 100 < 200" Range Count for write I/O operations reflects the number of one-second intervals during which the hIOmon I/O Monitor actually observed a total amount of data transferred by write I/O operations that was 100 000 000 bytes (i.e., 100 megabytes) or more, but less than 200 000 000 bytes (i.e., 200 megabytes).

The hIOmon "Performance Threshold Range Metrics" provide a quick and easy way to assess and characterize the actual "speeds and feeds" I/O performance of your particular files, devices and applications in everyday, normal usage!

The hIOmon "I/O Trender Display" also supports the display of the hIOmon "DataTransferred/Time Index (DXTI)" metric, which provides a "high-level" means for relative comparison of I/O performance, where basically "higher is better"; that is, the "higher" the Index number, the better the performance - as in more data transferred and/or less required response time (i.e., application wait).  It resembles the "fuel economy" index for an automotive vehicle (i.e., "miles-per-gallon" or "kilometres/litre") as an overall measure of "performance efficiency".

The DXTI metric is calculated by simply taking the observed overall total amount of data transferred by the respective I/O operations (converted to megabytes for scaling), and then dividing this combined total amount by the corresponding combined sum of the observed time durations (i.e., response times) of these I/O operations that were performed to transfer this data.  The resulting value is considered to be an index.

The hIOmon "I/O Trender Display" application runs in conjunction with the hIOmon "Disk I/O Ranger Display" application, which transfers to the I/O Trender Display various metrics that it has retrieved.  These metrics are then displayed by the I/O Trender Display in accordance with the various configuration options provided by the I/O Trender Display application.

See the help information provided by the hIOmon Disk I/O Ranger Display application for additional information about the hIOmon "Disk I/O Ranger Display".

Configuration Information

The hIOmon I/O Trender Display application is dependent upon the hIOmon Disk I/O Ranger Display as the source of the metrics displayed by the I/O Trender Display.  Consequently, the hIOmon Disk I/O Ranger Display application must be properly configured and running for the hIOmon I/O Trender Display application to be used.  See the help information provided by the hIOmon Disk I/O Ranger Display application for specific details about configuring and retrieving metrics using the hIOmon Disk I/O Ranger Display. 

Also please note that the hIOmon I/O Trender Display application provides its own configuration options that you can use to select, for example, the particular metrics that it is to display and how they are to be displayed.  See the Customize Display information below to find out more about the specific configuration options provided by the I/O Trender Display application. 

Installation Information

The hIOmon “I/O Trender Display” application is installed as part of the hIOmon "Disk I/O Ranger Display" application.  Specific files associated with the hIOmon "I/O Trender Display" application can be found in the “hIOmonDiskIORanger\hIOmonIOtrender” folder within the “..\hyperIO\hIOmon\Support\AddOns\” directory where the hIOmon software was installed; this applies to both the "hIOmon Workload I/O Profiler (WIOP)" and the "hIOmon Workload I/O Profiler Files (WIOPF)" software packages.

Please note that the hIOmon "I/O Trender" support must be enabled within the "hIOmon Disk I/O Ranger Display” in order to use the "I/O Trender Display".  A "Trender Display" button will appear next to the "Settings" button at the bottom of the main hIOmon "Disk I/O Ranger Display" window if the hIOmon "I/O Trender" support has been enabled and the required Internet Explorer version 10.0 or higher is installed.

Getting Started

To use/run the hIOmon "I/O Trender Display" application, simply click upon the "Trender Display” button that appears next to the "Settings" button at the bottom of the main hIOmon Disk I/O Ranger Display window (top frame).  The I/O Trender Display will then appear within the bottom frame of the “hIOmon Disk I/O Ranger Display” application main display.

The hIOmon I/O Trender Display application is initially configured so that only the combined "Read/Write Response Time (RT)" metrics are displayed.

Please note that this initial configuration for the hIOmon I/O Trender Display application can be changed by using the "Customize" button that appears at the top of the "I/O Trender Display".

Main Display

The Main Display of the hIOmon I/O Trender includes the graphical display of various hIOmon "Performance Threshold Range Metric" groups along with important contextual information.

The "Performance Threshold Range Metric" groups that can be optionally configured for display are the IOPS, MBs, Response Time (RT), Queue Depth (QD), and Data Transfer Size (DTS) ranges.  The range values are displayed within a "stacked format" using the left-side y-axis.  In the case of IOPS and MBs, this y-axis represents the percentage of time (e.g., the overall collection period).  For RT, QD, and DTS ranges, the y-axis represents the percentage of the overall combined total of the respective I/O operations.

Additional metrics such as the I/O operation counts, data transferred amounts, read/write ratios, and random/sequential access counts (both in terms of I/O operation counts and the respective amount of data transferred) can also be optionally displayed, with the right-side y-axis used.

A timestamp value (such the hour and minute) is shown at the bottom (i.e., the x-axis) of each graph displayed.

The primary items displayed include:

The "Settings" display of the hIOmon "Disk I/O Ranger Display" application can be used to select a different monitored item to be displayed.  Please refer to the Disk I/O Ranger Display help information for details regarding how to select a particular monitor item to be displayed.

The hIOmon read "DataTransferred/Time Index (DXTI)" metric is calculated by simply taking the observed overall total amount of data transferred by the applicable I/O operations (converted to megabytes for scaling), and then dividing this combined total amount by the corresponding combined sum of the observed time durations (i.e., response times) of the applicable I/O operations that were performed to transfer this data.  The resulting value is considered to be an index.

The basic idea behind the hIOmon "DataTransferred/Time Index" metric is straightforward and simple:  "Better" storage I/O operation performance is fundamentally about transferring (more) data faster.  So for a given amount of data, transferring this data more quickly (i.e., in less time).  And similarly, transferring more data within the same amount of time.  More specifically, the relationship is basically between the amount of data transferred and the corresponding amount of time that it took to perform the I/O operations required for the data transfer.

Overall, this Index metric provides a "high-level" means for relative comparison of I/O performance, where basically "higher is better"; that is, the "higher" the Index number, the better the performance - as in more data transferred and/or less required response time (i.e., application wait).  It resembles the "fuel economy" index for an automotive vehicle (i.e., "miles-per-gallon" or "kilometres/litre") as an overall measure of "performance efficiency".  It is akin to more miles driven (more data transferred) for fuel used (response time taken to transfer this data), or similarly, same miles (data transferred) but less fuel (less response time).

Each IOPS Range Count datapoint upon the graph represents the percentage of time (i.e., of the overall collection period) for the respective IOPS Range Count.  This percentage is the IOPS Range Count value divided by the total number of seconds that comprise the overall collection period.  As such, it basically indicates the percentage of time within the overall collection period that the respective IOPS Range Count value (e.g., "IOPS 1 - 99") was observed by the hIOmon I/O Monitor.  Use the tool-tip for a particular data point to display the specific percentage value for that IOPS Range along with the running total for all of the IOP Range Count values.

Each MBs Range Count datapoint upon the graph represents the percentage of time (i.e., of the overall collection period) for the respective MBs Range Count.  This percentage is the MBs Range Count value divided by the total number of seconds that comprise the overall collection period.  As such, it basically indicates the percentage of time within the overall collection period that the respective MBs Range Count value (e.g., "MBs < 1") was observed by the hIOmon I/O Monitor.  Use the tool-tip for a particular data point to display the specific percentage value for that MBs Range along with the running total for all of the MBs Range Count values.

Each RT Range Count datapoint upon the graph represents the respective RT Range Count value divided by the combined sum of the RT Range Count values.  As such, it basically indicates the percentage of applicable I/O operations that experienced the respective Response Time ranges as observed by the hIOmon I/O Monitor.  Use the tool-tip for a particular data point to display the specific percentage value for that RT Range along with the running total for all of the RT Range Count values.  The "running total" percentage reflects the percentage of applicable I/O operations that experienced a response time within the respective Response Time range or below (e.g., for the fourth Read RT Range Count whose range is "200 us < 500 us", the percentage of the total number of Read I/O operations whose response time was less than 500 microseconds).

Each QD Range Count datapoint upon the graph represents the respective QD Range Count value divided by the total number of I/O operations of the respective type.  As such, it basically indicates the percentage of applicable I/O operations that experienced the respective Queue Depth range as observed by the hIOmon I/O Monitor.  Use the tool-tip for a particular data point to display the specific percentage value for that QD Range along with the running total percentage for all of the QD Range Count values.  The "running total" percentage reflects the percentage of applicable I/O operations that experienced a queue depth within the respective Queue Depth range or below.

Please note that the "QD > 1" Range Count value reflects the accumulated number of times that the hIOmon I/O Monitor observed an I/O operations Queue Depth/Length that was greater than one; that is, where two or more I/O operations of the respective type were concurrently outstanding (had not yet completed and includes requests in service).  Also note that this particular Range Count value is displayed as an count (rather than a percentage as are the other QD Range Counts); the right-side y-axis shows count increments in this case.

Also please note the following:

  1. All of the above metrics reflect the total, cumulative values that have been automatically accumulated by the hIOmon I/O Monitor during the indicated overall collection period.  Note, however, that you can optionally display the metrics upon a "per-interval" basis.  Also please note that only those I/O operations that were successfully performed (i.e., transferred data) are included within the displayed IOPS, MBs, RT, and DTS Range Counts.

  2. Note that the first "one-second interval" for these Range Count metrics begins with the first monitored read (in the case of the read and the combined read/write metrics) or write (in the case of the write and the combined read/write metrics) I/O operation observed by the hIOmon I/O Monitor for the respective file, device, or process.

  3. The metrics are displayed upon a individual monitored item basis (i.e., for a single file, process, logical disk, physical volume, or physical device).  The "Settings" button/option provided by the hIOmon "Disk I/O Ranger Display" can be used to select a different monitored item for which the metrics are to be displayed.

  4. The display of the metric values is periodically refreshed/updated automatically to the current values retrieved by the hIOmon Disk I/O Ranger Display application.  You can view the refresh time interval by clicking on the "Settings" button/option in the hIOmon Disk I/O Ranger Display; these "Settings" options can also be used to select a different refresh time interval.  Also please note that only those IOPS, MBs, RT, QD, and DTS Range Counts that have a non-zero value are displayed.

  5. An error indication may be displayed near the bottom of the I/O Trender Display in the event that the I/O Trender Display or the Disk I/O Ranger Display application encounters an error.  Click on the hIOmon Disk I/O Ranger Display "Settings" option/button to view additional details about the error.

  6. On occasion, the tool-tip for a particular data point may fail to appear; this can, for instance, be due to the concurrent update of the I/O Trender Display frame.  To restore the tool-tips, refresh the I/O Trender Display by right-clicking the mouse and selecting the "Refresh" menu option.

  7. Up to three (3) separate graphs can be shown concurrently within the hIOmon I/O Trender Display.  These graphs/plots can be any combination of the IOPS, MBs, RT, QD, and DTS Range Counts for read, wriite, or combined read and write I/O operations.

Customize Display

The "Customize" button shown at the top of the hIOmon "I/O Trender Display" window can be used to optionally select various I/O Trender display and configuration options.  Click on this button to display the hIOmon I/O Trender Display "Settings" window, which provides a variety of selectable options including:

Please see the "I/O Count" display description above for additional details about this extended configuration option.

Please see the "Transfer Amount" display description above for additional details about this extended configuration option.

For example, if the "Minutes" option is selected and the collected metrics were collected/retrieved upon a 5 second basis (i.e., retrieved or summarized every 5 seconds), then the I/O Trender Display will display the metrics for each "one minute" sub-interval as shown in the heading of the top-most graph.  Please note in this case the "Show interval datapoints only" can also be selected so that the datapoints will be shown only for each sub-interval without additionally displaying each of the underlying datapoints (i.e., each of the 5-second datapoints in this example).

A "Show interval deltas" option is also provided.  You can use this option to display the metrics upon a per-interval basis rather than upon the total, cumulative basis, which is the default manner of display.  When this option is used, each data point reflects the change (e.g., increase or decrease) of the respective metric since the end of the preceding interval.

The "First DP Full Metrics" option enables you to display all of the metric values at the start of the initial interval.  By default, all metric values at the start of the initial period are generally zeroes.  However, there can be occasions (for instance, when the starting point for the display of an input log file is beyond the beginning of the log file) when the first interval displayed contains non-zero values.

The "Datapoint Limit" option can be used to limit the maximum number of datapoints shown across the time x-axis.

Also please note that the configuration changes are persistent.  That is, the configuration options will be saved so that they will automatically be used the next time that you start/invoke the hIOmon I/O Trender Display application.

Please note again that:

Error Messages

The following error, warning, and informational messages can be displayed by the hIOmon I/O Trender Display application:

Copyright © 2014-2019  hyperI/O LLC.  All Rights Reserved.