A Guide to Particle Advection Performance
Authors:
Abhishek Yenpure,
Sudhanshu Sane,
Roba Binyahib,
David Pugmire,
Christoph Garth,
Hank Childs
Abstract:
The performance of particle advection-based flow visualization techniques is complex, since computational work can vary based on many factors, including number of particles, duration, and mesh type. Further, while many approaches have been introduced to optimize performance, the efficacy of a given approach can be similarly complex. In this work, we seek to establish a guide for particle advection…
▽ More
The performance of particle advection-based flow visualization techniques is complex, since computational work can vary based on many factors, including number of particles, duration, and mesh type. Further, while many approaches have been introduced to optimize performance, the efficacy of a given approach can be similarly complex. In this work, we seek to establish a guide for particle advection performance by conducting a comprehensive survey of the area. We begin by identifying the building blocks for particle advection and establishing a simple cost model incorporating these building blocks. We then survey existing optimizations for particle advection, using two high-level categories: algorithmic optimizations and hardware efficiency. The sub-categories of algorithmic optimizations include solvers, cell locators, I/O efficiency, and precomputation, while the sub-categories of hardware efficiency all involve parallelism: shared-memory, distributed-memory, and hybrid. Finally, we conclude the survey by identifying current gaps in particle advection performance, and in particular on achieving a workflow for predicting performance under various optimizations.
△ Less
Submitted 20 January, 2022;
originally announced January 2022.
Scalable In Situ Lagrangian Flow Map Extraction: Demonstrating the Viability of a Communication-Free Model
Authors:
Sudhanshu Sane,
Abhishek Yenpure,
Roxana Bujack,
Matthew Larsen,
Kenneth Moreland,
Christoph Garth,
Hank Childs
Abstract:
We introduce and evaluate a new algorithm for the in situ extraction of Lagrangian flow maps, which we call Boundary Termination Optimization (BTO). Our approach is a communication-free model, requiring no message passing or synchronization between processes, improving scalability, thereby reducing overall execution time and alleviating the encumbrance placed on simulation codes from in situ proce…
▽ More
We introduce and evaluate a new algorithm for the in situ extraction of Lagrangian flow maps, which we call Boundary Termination Optimization (BTO). Our approach is a communication-free model, requiring no message passing or synchronization between processes, improving scalability, thereby reducing overall execution time and alleviating the encumbrance placed on simulation codes from in situ processing. We terminate particle integration at node boundaries and store only a subset of the flow map that would have been extracted by communicating particles across nodes, thus introducing an accuracy-performance tradeoff. We run experiments with as many as 2048 GPUs and with multiple simulation data sets. For the experiment configurations we consider, our findings demonstrate that our communication-free technique saves as much as 2x to 4x in execution time in situ, while staying nearly as accurate quantitatively and qualitatively as previous work. Most significantly, this study establishes the viability of approaching in situ Lagrangian flow map extraction using communication-free models in the future.
△ Less
Submitted 4 April, 2020;
originally announced April 2020.