You are here

Visualize This! Challenge

 

Table of Contents

What is the Visualize This! Challenge?

Now in its second year, Visualize This! is a Canada-wide competition that aims to celebrate the innovative ways visualization can help researchers explore datasets and answer important scientific questions. Visualize This! is your chance to challenge your creativity, experiment with new visualization tools, and contribute to the growth of data visualization in Canada!

This year’s Visualize This! Challenge is hosted by WestGrid, a regional partner of Compute Canada. Together with the Compute Canada Visualization Team, we want to bring your creative visualization ideas and workflows to life and make them accessible to all Canadian researchers!

To participate in the Challenge, please register your interest here.

This year’s contributed dataset is a multi-scale aerodynamical model of several counter-rotating vertical-axis wind turbines. It will be available for download from October 1 to November 30, 2017. Submissions are due no later than November 30 at midnight (PST).

Return to Table of Contents

Prizes

To be eligible for a prize, you must be registered as a participant. Register here!

These are just a few of the prizes we have up for grabs:

Return to Table of Contents

How to Participate

  • Register your interest here.
  • Download the dataset.
  • Final solutions must be submitted via email to alex.razoumov@westgrid.ca no later than November 30, 2017 (11:59 pm PST)
  • When you submit, please include both your visualization and the workflow, be it a script, a state file, etc. so that we can reproduce your solution.
  • Please do not attach large files (over 1 MB) to the email, but instead share them with us using any of the popular sharing tools such as Dropbox, Google Drive, ownCloud, or a shared cluster filesystem. Alternatively, you can upload your solution via anonymous ftp (write-only, no read access) “ftp blowfish.westgrid.ca (log in as anonymous, empty password); cd incoming; put yourLocalFile” or with a single command “curl -T yourLocalFile ftp://blowfish.westgrid.ca/incoming/ –user anonymous:”. With multiple files please pack and compress them into an archive before uploading.
  • Late submissions will not be considered.
Want tips for how to make your Visualize This entry stand out?

WestGrid is hosting a special online tutorial that will demonstrate how to create animations on static objects (zooming in, fly-troughs, spinning around, turning layers on/off) in both ParaView and VisIt.

Visualize This Webinar: Tips for Creating Animations
Wednesday, November 15
10:00 - 11:00 am PST 
Register Online Here

If you have questions about the challenge or your entry, bring them to this session and we'll answer them for you!

Return to Table of Contents

Why is Visualization Important?

Science is producing an unprecedented amount of data on everything from predicting weather and ocean changes to searching for distant galaxies. Visualization tools allow researchers to explore these large datasets interactively without being overwhelmed by the sheer amount of data. Ultimately, visualization enables researchers to answer the important scientific questions facing us today.

Return to Table of Contents

Download the Dataset

Click here to download the dataset as a 293MB compressed ZIP file.
If you have any issues accessing the dataset using that link, you can try this alternative link or email alex.razoumov@westgrid.ca for assistance. 

The dataset is taken from a simulation of the air flow around multiple counter-rotating vertical-axis wind turbines. It represents a single-time snapshot and contains the following five scalar variables on a 3D unstructured mesh:

  • pressure p relative to the reference pressure for normal conditions (can be positive or negative),
  • Q-criterion for vorticity q,
  • three Cartesian velocity components u, v, w

The original data produced by the simulation code was stored in Tecplot file format with explicit x,y,z coordinates for each of 5,307,199 mesh points and a list of 8 vertices for each of 16,679,253 cells. To compress the dataset and speed up I/O, we converted the data into a VTK file format, with the vertex coordinates and connections defining the cells encoded in the mesh itself. If for some reason you want data in the original Tecplot format, please let us know.

The main VTK file air.vtu stores field data (the five scalar variables) on a volumetric unstructured grid, whereas 10 smaller files blade{21,22,23,24,25,26,43,44,45,46}.vtp contain the field data on polygonal surface meshes defining the turbine blades.

The main challenge in visualizing this dataset is dealing with the range of scales from the entire simulation volume to small-scale turbulence near the blades.

Return to Table of Contents

How to Read the Data

Both ParaView and VisIt can open standard VTK unstructured grid and polygonal data files. These files can also be read in Python and C++ using the VTK library. Note that the volumetric file air.vtu contains over 16 million cells so it might take some time to read it, depending on your computer's speed.

As a result, rendering might also take some time, and can be automated with scripting, so that you could leave a script running for a few hours and come back to let's say several hundred frames of a movie, or it could be run in parallel on a cluster.

We hosted a kickoff webinar on Sept 27 that gave a walk-through of the dataset. Click here to view the archive recording (to get right to the dataset tour, skip ahead to 4:04 in the video).

Return to Table of Contents

Sample Visualizations

To give you an idea of the type of data in these files and to help you with actual visualizations, we provide two sample visualizations, one done with ParaView and the other one with VisIt. Both workflows demonstrate loading of all 11 VTK files.

The ParaView state file bladesWithLines.pvsm stores the pipeline to visualize the blades (coloured by the pressure on their surfaces) and the airflow around them with uniform-colour streamlines. You can point ParaView to this state file with File - Load State..., or start ParaView from the command line with "paraview --state=bladesWithLines.pvsm". The resulting image bladesWithLines.png is shown below.

turbine blades and airflow around them

The VisIt Python script positiveNegativePressure.py renders semi-transparent isosurfaces of positive (blue) and negative (turquoise) pressure around the blades. You can run this script in VisIt either from Controls - Launch CLI... or from Controls - Command..., or from the command line with "visit -nowin -cli -s positiveNegativePressure.py". The resulting image positiveNegativePressure0000.png is shown below.

isosurfaces of positive and negative pressure around the blades

We are looking for innovative visualizations of this dataset. For example, one could enhance these renderings by drawing streamlines around the isosurfaces and producing some animations such as spinning the visualization around the vertical axis or gradually turning on/off various visualization elements. Speaking more generally, a nice animation would help us explore the spatial range and values of multiple variables and show how various elements of the simulation are tied together.

Return to Table of Contents

Frequently Asked Questions (FAQ)

How do I register to participate in the Visualize This! Challenge?

Anyone interested in this competition must register their interest here. If you have any questions, please contact alex.razoumov@westgrid.ca.

Can we work in a team?

Yes, participants can form teams, but a team will be considered as a single entrant and will be eligible to win only one prize.

Is there a limit to how many people can be on a team?

No. There is no limit to the number of people who can be part of your team. It can be as large or as small as you like.

Do all team members need to register their interest?

No. Only one member of the team needs to register your interest in the challenge. Click here to register your interest.

What are the deadlines?

The dataset will be available for download from October 1 to November 30, with submissions due by midnight Pacific Time on November 30. Winners will be announced in December.

How will my application be judged?

Submissions will be reviewed by members of the Compute Canada Visualization Team. Points will be given for:

  • interactive visualizations,
  • visualizing the 3D multi-scale nature of the data,
  • online presentation in a browser, and
  • clear innovative display of multiple variables.

Who is eligible to participate?

The competition is open to anyone affiliated with a Canadian post-secondary institution (college or university) or research organization. Participants from all research fields are encouraged to enter. The competition dataset provides a valuable learning opportunity to enhance your visualization skills, no matter what your research area may be.

I have never used WestGrid or Compute Canada resources before. Can I still participate in the Challenge?

Yes. Your visualization does not need to be developed or implemented on WestGrid or Compute Canada's platforms. However, if you would like to use WestGrid or Compute Canada computational resources for your visualization, you can register for an account here: https://ccdb.computecanada.ca/security/login. If you have any questions about getting an account, please email accounts@computecanada.ca.

I am not affiliated with a Canadian university, post-secondary institution, or research organization. Can I participate in the Challenge?

Yes. If you partner with or join a team of someone who is affiliated with a Canadian university, post-secondary institution or research organization, you can participate in the Visualize This! Challenge.

I don’t live in Canada, but I have a WestGrid or Compute Canada account. Can I participate in the challenge?

Yes. If you have a WestGrid or Compute Canada account and are registered in the Compute Canada DataBase (CCDB), you are eligible to participate in the Visualize This! Challenge, even if you are not living in Canada.

What visualization tools can be used in the competition?

Participants can use any open-source visualization tool, including ParaView, VisIt, domain-specific packages, online Javascript/WebGL tools, Python visualization libraries, or any other open-source software you feel comfortable using. We hosted a kickoff webinar on Sept 27 that gave a walk-through of the dataset and gave some examples of visualization tools that can be used. Click here to view the archive recording (to get right to the dataset tour, skip ahead to 4:04 in the video).

Will any technical support be provided to participants?

If you have a specific technical question, please email alex.razoumov@westgrid.ca or bring it to our Visualize This! Tips & Tricks kickoff workshop on September 27. In general, we expect participants to create their own visualization(s), so technical help will be limited to answering very specific questions.

Our kickoff workshop will share more details on the dataset, supply sample routines to read the data, and demonstrate a few basic visualization examples. Click here to register for the workshop.

Who owns the Intellectual Property (IP) rights?

The dataset for this competition was kindly provided by a researcher and is his property or the property of his university. Participants own the rights to their visualization, but not to the original dataset. By submitting an application to the Visualize This! Challenge, the participant is providing consent for the use, reproduction, publication, transmission and/or broadcasting of his/her name, likeness, and/or prize information, and the solution, without compensation, in any publication or promotion by WestGrid or Compute Canada for the purposes of promoting the competition, the winners, and/or the benefits of visualization to research.

Can I use my own dataset?

No. All participants will work with the same dataset, which has been provided to the Challenge by a researcher. The dataset will be made available on the competition webpage and participants can download it anytime after October 01. We hosted a kickoff webinar on Sept 27 that gave a walk-through of the dataset. Click here to view the archive recording (to get right to the dataset tour, skip ahead to 4:04 in the video).

How do I submit my solution?

Solutions must be submitted via email to alex.razoumov@westgrid.ca.

Please follow these instructions when submitting:

  • Submit both your visualization and the workflow, be it a script, a state file, etc. so that we can reproduce your solution.
  • If you have any large files (over 1 MB), do not attach them to the email. Instead, share them with us using any of the popular sharing tools such as Dropbox, Google Drive, ownCloud, or a shared cluster filesystem. Alternatively, you can upload your solution via anonymous ftp (write-only, no read access) "ftp blowfish.westgrid.ca (log in as anonymous, empty password); cd incoming; put yourLocalFile" or with a single command "curl -T yourLocalFile ftp://blowfish.westgrid.ca/incoming/ --user anonymous:". With multiple files please pack and compress them into an archive before uploading.

When do I submit my solution?

Participants can submit their solutions at any time during the competition between October 1, 2017 (12:01 am PST) and November 30, 2017 (11:59 pm PST). Late submissions will not be considered.

Can I submit more than once?

Yes. Each submission will be time-stamped and the last submission received before the deadline date/time from each participant will be what is considered the final for submission.