# Rubik: A Hierarchical Architecture for Efficient Graph Learning

arXiv: https://arxiv.org/abs/2009.12495 · HTML: https://mlsanity.com/paper/2009.12495/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 26 Sep 2020.

## Built on by (most cited first, 7 of 7)

- [I-GCN: A Graph Convolutional Network Accelerator with Runtime Locality Enhancement through Islandization](https://mlsanity.com/paper/2203.03606/) (2022)
- [Do We Need Anisotropic Graph Neural Networks?](https://mlsanity.com/paper/2104.01481/) (2021)
- [GCoD: Graph Convolutional Network Acceleration via Dedicated Algorithm and Accelerator Co-Design](https://mlsanity.com/paper/2112.11594/) (2021)
- [GNNear: Accelerating Full-Batch Training of Graph Neural Networks with Near-Memory Processing](https://mlsanity.com/paper/2111.00680/) (2021)
- [LW-GCN: A Lightweight FPGA-based Graph Convolutional Network Accelerator](https://mlsanity.com/paper/2111.03184/) (2021)
- [Slice-and-Forge: Making Better Use of Caches for Graph Convolutional Network Accelerators](https://mlsanity.com/paper/2301.09813/) (2023)
- [COIN: Communication-Aware In-Memory Acceleration for Graph Convolutional Networks](https://mlsanity.com/paper/2205.07311/) (2022)

## Abstract (arXiv)

Graph convolutional network (GCN) emerges as a promising direction to learn the inductive representation in graph data commonly used in widespread applications, such as E-commerce, social networks, and knowledge graphs. However, learning from graphs is non-trivial because of its mixed computation model involving both graph analytics and neural network computing. To this end, we decompose the GCN learning into two hierarchical paradigms: graph-level and node-level computing. Such a hierarchical paradigm facilitates the software and hardware accelerations for GCN learning. We propose a lightweight graph reordering methodology, incorporated with a GCN accelerator architecture that equips a customized cache design to fully utilize the graph-level data reuse. We also propose a mapping methodology aware of data reuse and task-level parallelism to handle various graphs inputs effectively. Results show that Rubik accelerator design improves energy efficiency by 26.3x to 1375.2x than GPU platforms across different datasets and GCN models.