Compressed Neighbour Lists for SPH,Computer Graphics Forum

当前位置： X-MOL 学术 › Comput. Graph. Forum › 论文详情

Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)

Compressed Neighbour Lists for SPH
Computer Graphics Forum ( IF 2.7 ) Pub Date : 2019-11-26 , DOI: 10.1111/cgf.13890
Stefan Band ₁ , Christoph Gissler _{1,

2} , Matthias Teschner ₁

Affiliation

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.

中文翻译：

SPH 的压缩邻居列表

我们提出了一种新的压缩方案来存储采用平滑粒子流体动力学 (SPH) 的迭代求解器的邻居列表。压缩方案的灵感来自 Stream VByte，但使用从数据到数据字节的非线性映射，可节省高达 87% 的内存。它是 Cell-Linked-List (CLL) 概念的一个新变体的一部分，其灵感来自于对细胞-粒子关系进行改进处理的紧凑散列。我们表明，由此产生的邻居搜索在速度和内存消耗方面优于紧凑散列。可以在使用 172 GB 内存的 24 核 PC 上处理具有多达 13 亿个 SPH 粒子的 Divergence-Free SPH (DFSPH) 场景。可以在具有 112 个内核和 880 GB RAM 的消息传递接口 (MPI) 环境中处理具有超过 70 亿个 SPH 粒子的场景。邻居搜索对于交互式应用程序也很有用。在 12 核 PC 上，可以在不到 40 毫秒的时间内计算出多达 20 万个粒子的 DFSPH 模拟步骤。

更新日期：2019-11-26

点击分享查看原文

点击收藏

公开下载

阅读更多本刊最新论文本刊介绍/投稿指南11