NVIDIA IndeX accelerated computing for visualizing Cholla's galactic winds

Galactic winds – outflows of gas driven out of galaxies by the combined effects of thousands of supernovae – are a crucial feature of galaxy evolution. By removing gas from galaxies, they regulate future star formation, and distribute the dust and heavy elements formed in stars throughout the Universe. Despite their importance, a complete theoretical picture of these winds has been elusive. Simulating the complicated interaction between the hot, high pressure gas created by supernovae and the cooler, high density gas in the galaxy disk requires massive computational resources and highly sophisticated software. In addition, galactic wind simulations generate terabytes of output posing additional challenges regarding the effective analysis of the simulated physical processes. In order to address those challenges, we present NVIDIA IndeX as a scalable framework to visualize the simulation output. The framework features a streamingbased architecture to interactively explore simulation results in distributed multi-GPU environments. We demonstrate how to customize specialized sampling programs for volume and surface rendering to cover specific analysis questions of galactic wind simulations. This provides an extensive level of control over the visualization while efficiently using available resources to achieve high levels of performance and visual accuracy.

[1]  Markus Hadwiger,et al.  State‐of‐the‐Art in GPU‐Based Large‐Scale Volume Visualization , 2015, Comput. Graph. Forum.

[2]  B. Robertson,et al.  The Physical Nature of Starburst-driven Galactic Outflows , 2020, The Astrophysical Journal.

[3]  B. Robertson,et al.  Introducing CGOLS: The Cholla Galactic Outflow Simulation Suite , 2018, The Astrophysical Journal.

[4]  B. Robertson,et al.  Production of Cool Gas in Thermally Driven Outflows , 2018, The Astrophysical Journal.

[5]  Pat Hanrahan,et al.  Volume Rendering , 2020, Definitions.

[6]  Marc Levoy,et al.  Efficient ray tracing of volume data , 1990, TOGS.

[7]  B. Robertson,et al.  HYDRODYNAMICAL COUPLING OF MASS AND MOMENTUM IN MULTIPHASE GALACTIC WINDS , 2016, 1607.01788.

[8]  Evan E. Schneider,et al.  CHOLLA: A NEW MASSIVELY PARALLEL HYDRODYNAMICS CODE FOR ASTROPHYSICAL SIMULATION , 2014, 1410.4194.

[9]  Jens H. Krüger,et al.  An analysis of scalable GPU-based ray-guided volume rendering , 2013, 2013 IEEE Symposium on Large-Scale Data Analysis and Visualization (LDAV).

[10]  Christoph Garth,et al.  A Task-Based Parallel Rendering Component For Large-Scale Visualization Applications , 2017, EGPGV@EuroVis.

[11]  Markus Hadwiger,et al.  Real‐Time Ray‐Casting and Advanced Shading of Discrete Isosurfaces , 2005, Comput. Graph. Forum.

[12]  Charles Hansen,et al.  TOD-Tree: Task-Overlapped Direct Send Tree Image Compositing for Hybrid MPI Parallelism and GPUs , 2017, IEEE Transactions on Visualization and Computer Graphics.

[13]  E. Turner,et al.  The mass-to-light ratio of late-type binary galaxies - Luminosity- versus number-weighted averages , 1977 .