# Fast Stencil-Code Computation on a Wafer-Scale Processor

arXiv: https://arxiv.org/abs/2010.03660 · HTML: https://mlsanity.com/paper/2010.03660/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 7 Oct 2020.

## Built on by (most cited first, 10 of 10)

- [Graph Neural Networks in Particle Physics: Implementations, Innovations, and Challenges](https://mlsanity.com/paper/2203.12852/) (2022)
- [Demystifying AI Platform Design for Distributed Inference of Next-Generation LLM models](https://mlsanity.com/paper/2406.01698/) (2024)
- [VEGETA: Vertically-Integrated Extensions for Sparse/Dense GEMM Tile Acceleration on CPUs](https://mlsanity.com/paper/2302.08687/) (2023)
- [Bridging Data Center AI Systems with Edge Computing for Actionable Information Retrieval](https://mlsanity.com/paper/2105.13967/) (2021)
- [A Tale of Two Models: Constructing Evasive Attacks on Edge Models](https://mlsanity.com/paper/2204.10933/) (2022)
- [Applications of physics-informed scientific machine learning in subsurface science: A survey](https://mlsanity.com/paper/2104.04764/) (2021)
- [Evaluating Spatial Accelerator Architectures with Tiled Matrix-Matrix Multiplication](https://mlsanity.com/paper/2106.10499/) (2021)
- [An Experimental Evaluation of Machine Learning Training on a Real Processing-in-Memory System](https://mlsanity.com/paper/2207.07886/) (2022)
- [Implementation of a Binary Neural Network on a Passive Array of Magnetic Tunnel Junctions](https://mlsanity.com/paper/2112.09159/) (2021)
- [LoopTune: Optimizing Tensor Computations with Reinforcement Learning](https://mlsanity.com/paper/2309.01825/) (2023)

## Abstract (arXiv)

The performance of CPU-based and GPU-based systems is often low for PDE codes, where large, sparse, and often structured systems of linear equations must be solved. Iterative solvers are limited by data movement, both between caches and memory and between nodes. Here we describe the solution of such systems of equations on the Cerebras Systems CS-1, a wafer-scale processor that has the memory bandwidth and communication latency to perform well. We achieve 0.86 PFLOPS on a single wafer-scale system for the solution by BiCGStab of a linear system arising from a 7-point finite difference stencil on a 600 X 595 X 1536 mesh, achieving about one third of the machine's peak performance. We explain the system, its architecture and programming, and its performance on this problem and related problems. We discuss issues of memory capacity and floating point precision. We outline plans to extend this work towards full applications.