IT Operations Management (ITOM)

vPV, tree maps and the Mantra* of visual analytics

vPV, tree maps and the Mantra* of visual analytics

Ramkumar Devana

* Mantra is an often repeated word, formula, or phrase, often a truism - definition


(This blog is written with inputs from Goranka Medhi, Product Architect, HP Software)



“Overview first, zoom and filter, then details-on-demand.”

Ben Shneiderman, 1996



The tree-map, used in tools such as HP’s vPV ( is built on a space-constrained visualization approach in order to display a large set of objects and their status in a visual pane.



(click to view image in full-size)


The idea is to be able to show the size and colour of the ‘leaf’ – object or objects under consideration represented in the tree-map however large the environment under study, such as files on a disk, or stocks in a stock market.


(click to view image in full-size)


The earliest use of tree-maps is by Ben Shneiderman (see - the inventor of the tree-map idea. His idea was to build a representation to view files on an easily-filled disk and figure out the biggest files in this. So he built a tool which after multiple adaptations is available today as the TreeViz application free for Mac users.


The most important outcome of the tree map representation is spotting patterns of problems across entire groups of elements.


“Showing the entire data set presents the opportunity to identify emergent patterns – characteristics that are very difficult to express analytically but trivially identified visually.”

Martin Krzywinski,




Visual analytics

"the science of analytical reasoning facilitated by visual interactive interfaces."

James J. Thomas and Kristin A. Cook (Ed.) (2005)


When using online maps, how many times have you printed an entire map, along with the directions to a place…?? The human perceptive ability handles pictures better than it does words or numbers.


The science of visual analytics deals with using a graphical representation of data to enable better analytics. A picture can compensate for 1000 words and similarly a visual way of representation generally tends to offer quicker way to resolving problems (or at least a quicker way to identifying the problem/pattern).


The unique thing about the tree-map in vPV is it brings in additional capabilities consistent with Shneiderman’s visual information seeking mantra of starting with the overview, allowing filtering/grouping and then bringing up details on demand. The idea is to assess how things are behaving overall in a virtualized environment, then go into ‘hot’ zones and drill-down into the problem using temporal, comparative and aggregative approaches – thoughts for another blog.


Here’s an example of how vPV helps to isolate a problem pattern to a particular cluster, thereby hastening analysis and potential resolution of the problem.


(click to view image in full-size)


Notice in image above, how after grouping all VMs encountering a high cpu-ready utilization now appear to run within a VMware cluster. This is where the job of analysis and further troubleshooting can further be done using vPV’s workbench capabilities and drill-down to guest OS process metrics.


(click to view image in full-size) 


In the image above, there are several VMs with high I/O latency which seems to be occurring all over the board. See how these problems become isolated ones, just by looking at the grouping/container they belong in.


The video below shows a large VMware setup being monitored using vPV. In this setup, how does the vAdmin arrive at the most crucial VMs and datastores suffering from IOPS issues - have a look at the video (shameless plug!).






A word of note for users of traditional software tools


Most traditional users of software tools have had more experience seeing rows and rows of data in tables or reports or event consoles, and less experience with visual representation.


There is a learning curve here of course that folks new to this will need to but the benefits of going through this experience will help them to get a far quicker view of the real problems in their environment. The pic below shows how a user filters the vPV view to see only those VMs with higher than 50% utilization.


(click to view image in full-size) 



Tree-map used in other virtualization and cloud monitoring tools


The tree-map is a general visual representation and so it can be used in various fields of study for differing uses/users. Many operations or other monitoring tools use tree-maps to convey the overview picture of a monitored environment. Of interest is a tree-map used in VMware’s vCenter Operations Manager’s capacity reports (see image below).



Here’s an image showing tree-map kind of visualization from BMC’s Operations Manager tool.



In all, there’s a high amount of interest and relevance for visual analytics in the tools used by IT practitioners and managers.



Other links to look into


Some other tree map applications and other nice visualization techniques that are cute as well as are functional are –


  1. HP’s Network Flow Analytics - free tool uses the tree-map representation that vPV uses. For details, go to
  2. Smart Money’s Map of the Market -
  3. A tree-map representation of aggregated news articles – ‘NewsMap’ @
  4. Cubism.js ( for visualization of  time-series data of multiple parallel but related object sets
  5. Bullet charts ( or thermometer controls (also used in HP SHO’s utilization analysis charts - see image below)
  6. Some great visualization and infographics @
  7. SiteScope also now provides a new tree-map in their Multi-view feature - to read more about it, Multi-View - A new perspective for displaying the status of all monitors in a single view








Disclaimer on copyrights -

All images used above for are as obtained directly from product documents or websites. The intent is to use these images only as a reference study.


Disclaimer on links in this article -

Links in this blog may lead to external blogs. HP is not responsible for the content.

HPE Software Rocks!
  • infrastructure management
About the Author

Ramkumar Devana

Ramkumar Devanathan (twitter: @rdevanathan) is Product Manager for HPE Cloud Optimizer (formerly vPV). He was previously a member of the IOM-Customer Assist Team (CAT) providing technical assistance to HP Software pre-sales and support teams with Operations Management products including vPV, SHO, VISPI. He has experience of more than 14 years in this product line, working in various roles ranging from developer to product architect.