Data Analysis Services Group - January 2014

News and Accomplishments

VAPOR Project

Project information is available at: http://www.vapor.ucar.edu

KISTI Award:
The team prepared to submit a Statement of Work for the KISTI Year Four project, scoping out the effort of various activities requested by KISTI:

  • Scott and John familiarized themselves with GRIMs and CAM data.  This entailed gathering and reviewing documentation on GRIB as well as GRIMs, the KISTI model that outputs GRIB data.  Additionally, time was spent exploring analysis tools including Metview, NCL, wGRIB, and the ECMWF GRIB_API.
  • Alan prepared an estimate for implementing an isoline renderer.

VAPOR 2.x Development:


The team continued to fix bugs in the 2.3.0 release candidate in the hope of releasing a stable version of the code in early February

  • Scott resolved a performance inconsistency between the vdcwizard GUI and the command line data conversion tools (the latter was significantly slower)
  • John developed a class wrapper for the Proj4 API in an effort to provide a single, and consistent method for handling geo-referenced data throughout the code base. In the process numerous long-standing geo-referencing related bugs were fixed.
  • John made an attempt to fix clipping in the isosurface renderer (currently a no-op) but abandoned the effort when the scope of changes grew too large.
  • Alan fixed a few bugs, including improved probe cropping and two animation problems.
  • Rick Brownrigg fixed an NCL bug (in wrf2geotiff.ncl).

Scott started exploring more general support for user annotation,  but set the work aside since our meeting on vaporgui code for 3.0, since glwindow.cpp will have an incompatible implementation (no QT).

Miles continued work on a class object for rendering flow primitives (streamlines, vectors, points).

VAPOR 3.x Development:

The group resumed work on 3.0 development, starting with a kick-off meeting in early January to discuss plans. A second meeting was held to review a 3.0 GUI prototype that it is hope will eventually serve as a template for production development.


Documentation:

During our team meeting on Jan. 6, we discussed the prospect of using YouTube videos in our documentation.  It was agreed that online videos for Vapor would be a powerful supplement to the website, but the concept never gained inertia due to competing priorities. Scott saw this as an excellent opportunity to 1) contribute to our documentation, 2) demonstrate some of my skillset in digital media, and 3) showcase ffmpeg, which is a cross platform media encoder, licensed under the GPL.  It may be a good tool for incorporating a movie-generator into Vapor.

Scott made a proof-of-concept tutorial video and published it onto youtube at the following url:https://www.youtube.com/watch?v=5-TNjl_9IdY

B.J. Smith is working with Alan to provide a short introduction to VAPOR for Yellowstone users.


Administrative:

Alan and Scott reviewed applicants for Sciparcs 2014 and selected our top three candidates.  Before we could pick a student, we had to elicit the requirements for the intern’s project.  Once we had a clearer objective, we agreed on who we thought were the best fits.

John prepared slides for the OaPM meeting on DAV metrics.

Tim Scheitlin is in the process of hiring a software engineer to do video editing and more.  Alan has been helping interview candidates for this position.

Education and Outreach:

Alan and Scott provided VAPOR support for attendees at the WRF tutorial labs on Jan 30 and 31.


Software Research Projects


A pre-proposal for the DOE's Exascale Visualization FOA was submitted in partnership with Kenny Gruchalla (NREL) and Hank Childs (U. of Oregon)

Feature Tracking:

Climate data compression:

Production Visualization Services & Consulting

 

Alan worked with Mel Shapiro, helping him prepare for his AMS2014 presentations of Sandy and ERICA visualizations.

 

Scott met an HAO physicist named Hanli Liu.  Hanli works with CAM data, modeling microbaroms and gravity waves, primarily in the upper atmosphere. He was very interested in VAPOR so Scott has been visualizing his data using our conversion tools.  By pure coincidence, KISTI also seems to have interest in having us convert CAM data, so his model output might be a good test platform.

ASD Support

  • xxx

Publications, Papers & Presentations

Yuk, J., Clyne, J., Norton, A., and Joh, M. Enhancement of VAPOR to Support Ocean Data, Journal of Supercomputing Information, Vol. 1, No. 2, December 2013, 15-21

Professional development

 

Scott attended the NCL workshop from Jan. 14-17.  While learning about the language syntax and 2D plotting features, Scott also used it as an opportunity to learn more about the GRIB data format with a sample of JinHee’s GRIMs model output.


 

Publications, Papers & Presentations


Data Services

  • xxx

Accounting & Statistics

  • xxx

Security & Administration

  • xxx

System Monitoring

  • xxx

System Support

ML - Data Analysis & Visualization Clusters

  • xxx

GLADE Storage Cluster

  • xxx

Data Transfer Cluster

  • xxx

Experimental Clusters

  • xxx

Test Clusters

Storage Usage Statistics

NWSC+GLADE+Usage+Report

Other

  • xxx
  • No labels