Packet-Oriented Streamline Tracing on Modern SIMD Architectures

The advection of integral lines is an important computational kernel in vector field visualization. We investigate how this kernel can profit from vector (SIMD) extensions in modern CPUs. As a baseline, we formulate a streamline tracing algorithm that facilitates auto-vectorization by an optimizing compiler. We analyze this algorithm and propose two different optimizations. Our results show that particle tracing does not per se benefit from SIMD computation. Based on a careful analysis of the auto-vectorized code, we propose an optimized data access routine and a re-packing scheme which increases average SIMD efficiency. We evaluate our approach on three different, turbulent flow fields. Our optimized approaches increase integration performance up to 5.6x over our baseline measurement. We conclude with a discussion of current limitations and aspects for future work.

[1]  Rüdiger Westermann,et al.  Interactive Separating Streak Surfaces , 2010, IEEE Transactions on Visualization and Computer Graphics.

[2]  Kellogg S. Booth,et al.  Report from the chair , 1986 .

[3]  Filip Sadlo,et al.  Efficient Visualization of Lagrangian Coherent Structures by Filtered AMR Ridge Extraction , 2007, IEEE Transactions on Visualization and Computer Graphics.

[4]  Kenneth I. Joy,et al.  GPU Acceleration of Particle Advection Workloads in a Parallel, Distributed Memory Setting , 2013, EGPGV@Eurographics.

[5]  Rüdiger Westermann,et al.  A particle system for interactive visualization of 3D flows , 2005, IEEE Transactions on Visualization and Computer Graphics.

[6]  Han-Wei Shen,et al.  Load-Balanced Parallel Streamline Generation on Large Scale Vector Fields , 2011, IEEE Transactions on Visualization and Computer Graphics.

[7]  G. Haller.  Distinguished material surfaces and coherent structures in three-dimensional fluid flows , 2001 .

[8]  H. Hagen,et al.  Interactive Ray Tracing of Arbitrary Implicits with SIMD Interval Arithmetic , 2007, 2007 IEEE Symposium on Interactive Ray Tracing.

[9]  Kenneth I. Joy,et al.  Evaluating the benefits of an extended memory hierarchy for parallel streamline algorithms , 2011, 2011 IEEE Symposium on Large Data Analysis and Visualization.

[10]  Jens Schneider,et al.  Interactive Visual Exploration of Unsteady 3D Flows , 2007, EuroVis.

[11]  Tom Peterka,et al.  Parallel particle advection and FTLE computation for time-varying flow fields , 2012, 2012 International Conference for High Performance Computing, Networking, Storage and Analysis.

[12]  Kenneth I. Joy,et al.  Streamline Integration Using MPI-Hybrid Parallelism on a Large Multicore Architecture , 2011, IEEE Transactions on Visualization and Computer Graphics.

[13]  Ingo Wald,et al.  Combining Single and Packet-Ray Tracing for Arbitrary Ray Distributions on the Intel MIC Architecture , 2012, IEEE Transactions on Visualization and Computer Graphics.

[14]  Hans Hagen,et al.  Full-resolution interactive CPU volume rendering with coherent BVH traversal , 2011, 2011 IEEE Pacific Visualization Symposium.

[15]  James P. Ahrens,et al.  PISTON: A Portable Cross-Platform Framework for Data-Parallel Visualization Operators , 2012, EGPGV@Eurographics.

[16]  Ingo Wald,et al.  Embree: a kernel framework for efficient CPU ray tracing , 2014, ACM Trans. Graph..

[17]  Maya Gokhale,et al.  Multi-threaded streamline tracing for data-intensive architectures , 2014, 2014 IEEE 4th Symposium on Large Data Analysis and Visualization (LDAV).

[18]  Lipo Wang,et al.  The length-scale distribution function of the distance between extremal points in passive scalar turbulence , 2006, Journal of Fluid Mechanics.

[19]  Kwan-Liu Ma,et al.  Dax Toolkit: A proposed framework for data analysis and visualization at Extreme Scale , 2011, 2011 IEEE Symposium on Large Data Analysis and Visualization.

[20]  Gunther H. Weber,et al.  Scalable computation of streamlines on very large datasets , 2009, Proceedings of the Conference on High Performance Computing Networking, Storage and Analysis.

[21]  Norbert Peters,et al.  The vorticity versus the scalar criterion for the detection of the turbulent/non-turbulent interface , 2014, Journal of Fluid Mechanics.

[22]  Robert Sisneros,et al.  EAVL: The Extreme-scale Analysis and Visualization Library , 2012, EGPGV@Eurographics.

[23]  Torsten Kuhlen,et al.  Virtual Tubelets - efficiently visualizing large amounts of particle trajectories , 2005, Comput. Graph..

[24]  S. Boulos,et al.  Getting rid of packets - Efficient SIMD single-ray traversal using multi-branching BVHs - , 2008, 2008 IEEE Symposium on Interactive Ray Tracing.

[25]  Daniel Weiskopf,et al.  Hierarchical Line Integration , 2011, IEEE Transactions on Visualization and Computer Graphics.

[26]  Torsten Kuhlen,et al.  Interactive particle tracing in time-varying tetrahedral grids , 2011, EGPGV '11.

[27]  Brian Cabral,et al.  Imaging vector fields using line integral convolution , 1993, SIGGRAPH.

[28]  Min Chen,et al.  Over Two Decades of Integration‐Based, Geometric Flow Visualization , 2010, Comput. Graph. Forum.

[29]  Hans Hagen,et al.  Efficient Computation and Visualization of Coherent Structures in Fluid Flow Applications , 2007, IEEE Transactions on Visualization and Computer Graphics.

[30]  Kenneth I. Joy,et al.  Generation of Accurate Integral Surfaces in Time-Dependent Vector Fields , 2008, IEEE Transactions on Visualization and Computer Graphics.

[31]  Kenneth I. Joy,et al.  Fast, Memory-Efficient Cell Location in Unstructured Grids for Visualization , 2010, IEEE Transactions on Visualization and Computer Graphics.

[32]  Jeff P. Hultquist,et al.  Constructing stream surfaces in steady 3D vector fields , 1992, Proceedings Visualization '92.