Compressed Neighbour Lists for SPH. (26th November 2019)

• Record Type: Journal Article
• Authors: Band, Stefan; Gissler, Christoph; Teschner, Matthias
• Is Part Of: Computer graphics forum. Volume 39:Number 1(2020)
• Page Start: 531
• Page End: 542
• Publication Date: 2019-11-26
• Subjects: physically based animation -- data compression -- fluid modelling -- smoothed particle hydrodynamics -- I.3.7 [Computer Graphics]: Three‐Dimensional Graphics and Realism – Animation; Computing methodologies → Physical simulation -- Massively parallel and high‐performance simulations; Computer graphics -- Periodicals; 006.605
• DOI: 10.1111/cgf.13890
• Languages: English
• ISSNs: 0167-7055

Abstract

Abstract: We propose a novel compression scheme to store neighbour lists for iterative solvers that employ Smoothed Particle Hydrodynamics (SPH). The compression scheme is inspired by Stream VByte, but uses a non‐linear mapping from data to data bytes, yielding memory savings of up to 87%. It is part of a novel variant of the Cell‐Linked‐List (CLL) concept that is inspired by compact hashing with an improved processing of the cell‐particle relations. We show that the resulting neighbour search outperforms compact hashing in terms of speed and memory consumption. Divergence‐Free SPH (DFSPH) scenarios with up to 1.3 billion SPH particles can be processed on a 24‐core PC using 172 GB of memory. Scenes with more than 7 billion SPH particles can be processed in a Message Passing Interface (MPI) environment with 112 cores and 880 GB of RAM. The neighbour search is also useful for interactive applications. A DFSPH simulation step for up to 0.2 million particles can be computed in less than 40 ms on a 12‐core PC. Abstract : We propose a novel compression scheme to store neighbour lists for iterative solvers that employ Smoothed Particle Hydrodynamics (SPH). The compression scheme is inspired by Stream VByte, but uses a non‐linear mapping from data to data bytes, yielding memory savings of up to 87%. It is part of a novel variant of the Cell‐Linked‐List (CLL) concept that is inspired by compact hashing with an improved processing of the cell‐particle relations. We show that the resulting neighbour search outperforms compact hashing in terms of speed and memory consumption. Divergence‐Free SPH (DFSPH) scenarios with up to 1.3 billion SPH particles can be processed on a 24‐core PC using 172 GB of memory. Scenes with more than 7 billion SPH particles can be processed in a Message Passing Interface (MPI) environment with 112 cores and 880 GB of RAM. The neighbour search is also useful for interactive applications. A DFSPH simulation step for up to 0.2 million particles can be computed in less than 40 ms on a 12‐core PC.