将AI加速的流体仿真模型适配到IPU平台，实现16核下吞吐量提升5倍
Adaptation of AI-accelerated CFD Simulations to the IPU platform
- 用Poplar SDK定制TensorFlow，适配IPU-POD16平台训练流体仿真模型
- 通过popdist优化数据传输，单机提速34%；16核并行达2805.8样本/秒
- 适合关注高性能计算与AI融合的仿真研究者，尤其在多IPU协同场景
智能处理单元（IPU）已在诸多AI应用中展现价值。本文评估其在新兴的‘AI for simulation’领域中的表现，即利用人工智能支持传统数值模拟。重点针对一个用于训练支持计算流体动力学（CFD）应用的机器学习模型程序。使用Poplar SDK提供的定制TensorFlow，将程序适配至IPU-POD16平台，并研究其易用性与性能可扩展性。基于OpenFOAM仿真数据训练模型，可在测试阶段准确预测仿真状态。通过使用popdist库克服主机端向IPU输送训练数据的性能瓶颈，实现最高34%的加速。由于通信开销，采用数据并行同时使用两个IPU并未提升吞吐量。然而，在支付单个IPU内部计算成本后，其跨IPU通信硬件能力支持良好扩展性：从2个增至16个IPU，吞吐量由560.8提升至2805.8样本/秒。
原文摘要 · Abstract (English)
Intelligence Processing Units (IPU) have proven useful for many AI applications. In this paper, we evaluate them within the emerging field of \emph{AI for simulation}, where traditional numerical simulations are supported by artificial intelligence approaches. We focus specifically on a program for training machine learning models supporting a \emph{computational fluid dynamics} application. We use custom TensorFlow provided by the Poplar SDK to adapt the program for the IPU-POD16 platform and investigate its ease of use and performance scalability. Training a model on data from OpenFOAM simulations allows us to get accurate simulation state predictions in test time. We show how to utilize the \emph{popdist} library to overcome a performance bottleneck in feeding training data to the IPU on the host side, achieving up to 34\% speedup. Due to communication overheads, using data parallelism to utilize two IPUs instead of one does not improve the throughput. However, once the intra-IPU costs have been paid, the hardware capabilities for inter-IPU communication allow for good scalability. Increasing the number of IPUs from 2 to 16 improves the throughput from 560.8 to 2805.8 samples/s.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。