# GRANITE: A Graph Neural Network Model for Basic Block Throughput Estimation

arXiv: https://arxiv.org/abs/2210.03894 · HTML: https://mlsanity.com/paper/2210.03894/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- Builds on 31 of its 72 references that are in the map.
- First posted on arXiv 8 Oct 2022.

## Builds on (most cited first, 30 of 31)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Semi-Supervised Classification with Graph Convolutional Networks](https://mlsanity.com/paper/1609.02907/) (2016)
- [Layer Normalization](https://mlsanity.com/paper/1607.06450/) (2016)
- [Neural Message Passing for Quantum Chemistry](https://mlsanity.com/paper/1704.01212/) (2017)
- [A Comprehensive Survey on Graph Neural Networks](https://mlsanity.com/paper/1901.00596/) (2019)
- [Graph Neural Networks: A Review of Methods and Applications](https://mlsanity.com/paper/1812.08434/) (2018)
- [Relational inductive biases, deep learning, and graph networks](https://mlsanity.com/paper/1806.01261/) (2018)
- [Visualizing the Loss Landscape of Neural Nets](https://mlsanity.com/paper/1712.09913/) (2017)
- [In-Datacenter Performance Analysis of a Tensor Processing Unit](https://mlsanity.com/paper/1704.04760/) (2017)
- [Interaction Networks for Learning about Objects, Relations and Physics](https://mlsanity.com/paper/1612.00222/) (2016)
- [A simple neural network module for relational reasoning](https://mlsanity.com/paper/1706.01427/) (2017)
- [code2vec: Learning Distributed Representations of Code](https://mlsanity.com/paper/1803.09473/) (2018)
- [Learning to Represent Programs with Graphs](https://mlsanity.com/paper/1711.00740/) (2017)
- [End-to-End Relation Extraction using LSTMs on Sequences and Tree Structures](https://mlsanity.com/paper/1601.00770/) (2016)
- [Cross-Sentence N-ary Relation Extraction with Graph LSTMs](https://mlsanity.com/paper/1708.03743/) (2017)
- [Improved Code Summarization via a Graph Neural Network](https://mlsanity.com/paper/2004.02843/) (2020)
- [Graph Convolutional Encoders for Syntax-aware Neural Machine Translation](https://mlsanity.com/paper/1704.04675/) (2017)
- [Graph-to-Sequence Learning using Gated Graph Neural Networks](https://mlsanity.com/paper/1806.09835/) (2018)
- [Heterogeneity-Aware Cluster Scheduling Policies for Deep Learning Workloads](https://mlsanity.com/paper/2008.09213/) (2020)
- [Ithemal: Accurate, Portable and Fast Basic Block Throughput Estimation using Deep Neural Networks](https://mlsanity.com/paper/1808.07412/) (2018)
- [Estimating Node Importance in Knowledge Graphs Using Graph Neural Networks](https://mlsanity.com/paper/1905.08865/) (2019)
- [MLGO: a Machine Learning Guided Compiler Optimizations Framework](https://mlsanity.com/paper/2101.04808/) (2021)
- [An Evaluation of Edge TPU Accelerators for Convolutional Neural Networks](https://mlsanity.com/paper/2102.10423/) (2021)
- [Exploiting Semantics in Neural Machine Translation with Graph Convolutional Networks](https://mlsanity.com/paper/1804.08313/) (2018)
- [Data-Driven Offline Optimization For Architecting Hardware Accelerators](https://mlsanity.com/paper/2110.11346/) (2021)
- [Mind Mappings: Enabling Efficient Algorithm-Accelerator Mapping Space Search](https://mlsanity.com/paper/2103.01489/) (2021)
- [ProGraML: Graph-based Deep Learning for Program Optimization and Analysis](https://mlsanity.com/paper/2003.10536/) (2020)
- [Apollo: Transferable Architecture Exploration](https://mlsanity.com/paper/2102.01723/) (2021)
- [Learning Execution through Neural Code Fusion](https://mlsanity.com/paper/1906.07181/) (2019)

## Built on by (most cited first, 9 of 9)

- [TpuGraphs: A Performance Prediction Dataset on Large Tensor Computational Graphs](https://mlsanity.com/paper/2308.13490/) (2023)
- [LayerDAG: A Layerwise Autoregressive Diffusion Model for Directed Acyclic Graph Generation](https://mlsanity.com/paper/2411.02322/) (2024)
- [ArchGym: An Open-Source Gymnasium for Machine Learning Assisted Architecture Design](https://mlsanity.com/paper/2306.08888/) (2023)
- [Application Performance Modeling via Tensor Completion](https://mlsanity.com/paper/2210.10184/) (2022)
- [Tao: Re-Thinking DL-based Microarchitecture Simulation](https://mlsanity.com/paper/2404.10921/) (2024)
- [Learning Generalizable Program and Architecture Representations for Performance Modeling](https://mlsanity.com/paper/2310.16792/) (2023)
- [COMET: Neural Cost Model Explanation Framework](https://mlsanity.com/paper/2302.06836/) (2023)
- [A Benchmark on Directed Graph Representation Learning in Hardware Designs](https://mlsanity.com/paper/2410.06460/) (2024)
- [On the Limits of Machine-Learned Ranking for Modern Microarchitectural Policies](https://mlsanity.com/paper/2608.01041/) (2026)

## Abstract (arXiv)

Analytical hardware performance models yield swift estimation of desired hardware performance metrics. However, developing these analytical models for modern processors with sophisticated microarchitectures is an extremely laborious task and requires a firm understanding of target microarchitecture's internal structure. In this paper, we introduce GRANITE, a new machine learning model that estimates the throughput of basic blocks across different microarchitectures. GRANITE uses a graph representation of basic blocks that captures both structural and data dependencies between instructions. This representation is processed using a graph neural network that takes advantage of the relational information captured in the graph and learns a rich neural representation of the basic block that allows more precise throughput estimation. Our results establish a new state-of-the-art for basic block performance estimation with an average test error of 6.9% across a wide range of basic blocks and microarchitectures for the x86-64 target. Compared to recent work, this reduced the error by 1.7% while improving training and inference throughput by approximately 3.0x. In addition, we propose the use of multi-task learning with independent multi-layer feed forward decoder networks. Our results show that this technique further improves precision of all learned models while significantly reducing per-microarchitecture training costs. We perform an extensive set of ablation studies and comparisons with prior work, concluding a set of methods to achieve high accuracy for basic block performance estimation.