Nick Leaf, Venkatram Vishwanath, Joseph Insley, Mark Hereld, Michael E. Papka, and Kwan-Liu Ma have won the Best Paper Award at the 2013 IEEE Symposium on Large-Scale Data Analysis and Visualization (LDAV) for their paper “Efficient Parallel Volume Rendering of Large-Scale Adaptive Mesh Refinement Data.”
The paper was based on research sponsored in part by the National Science Foundation and the U.S. Department of Energy and was conducted using supercomputing resources at the Argonne Leadership Computing Facility (ALCF), a DOE leadership computing facility dedicated to open science located at Argonne National Laboratory.
Nick Leaf, a University of California, Davis graduate student, worked at Argonne with computer scientists Vishwanath, Insley, Hereld, and Papka, and his advisor and UC Davis computer science professor Kwan-Liu Ma, to develop a novel, cluster- and GPU-parallel rendering scheme for adaptive mesh refinement (AMR) data.
AMR is a popular approach for allocating scarce computing resources on large-scale supercomputers, such as memory, to the most important portions of the simulation domain. The team focused its efforts on FLASH, a multiscale multiphysics code used in numerous scientific domains, including astrophysics, high-energy-density physics, and cosmology.
“Volume rendering of AMR data is a very complex computer science problem,” said Vishwanath. “This is an effort to do visual analysis of AMR data generated by multiphysics codes running on today’s largest systems.”
As part of the ongoing effort, the team is developing scalable algorithms to visualize and analyze the data -- as the simulation is running -- to glean insights into the evolution and features of the simulation.
The LDAV Symposium is held in conjunction with IEEE VIS, an annual meeting of scientists, data analytics and visualization researchers, and users, to present the latest research innovations and to foster the exchanges needed to develop the next-generation of technologies. VIS 2013 is being held in Atlanta from October 13-18. The LDAV Best Paper Award was announced October 14.