让图神经网络在低功耗设备上跑得更快更省电。
GraNNite: Enabling High-Performance Execution of Graph Neural Networks on Resource-Constrained Neural Processing Units
- 分步优化：拆分任务、利用稀疏性、动态适应图结构。
- 比默认NPU快2.6至7.6倍，能效提升8.6倍。
- 适合边缘设备部署，尤其对实时推荐和隐私敏感场景有用。
图神经网络（GNN）在图数据学习中至关重要，广泛应用于网络分析、推荐系统与语音分析，并助力大语言模型的检索增强生成（RAG）及事件驱动视觉任务。然而，其不规则内存访问、稀疏性与动态结构导致在资源受限设备上延迟高、能耗大。尽管现代边缘处理器集成CPU、GPU与专用神经处理单元（NPU），但专为数据并行设计的NPU难以高效执行不规则的GNN计算。本文提出GraNNite，首个面向商用主流深度神经网络加速器的硬件感知框架，采用三步法：（1）实现NPU执行，通过GraphSplit分发负载、StaGr静态聚合，以及GrAd与NodePad处理动态图；（2）性能优化，采用EffOp应对控制密集型任务，GraSp挖掘稀疏性，预计算优化PreG、SymG、CacheG减少冗余与内存传输；（3）权衡精度与效率，引入INT8量化QuantGr，以及GrAx1/2/3分别加速注意力、广播加法与SAGE-max聚合。在Intel Core Ultra AI PC上，GraNNite相较默认NPU映射提速2.6–7.6倍，相比CPU/GPU最高节能8.6倍，性能分别达其10.8倍与6.7倍。
原文摘要 · Abstract (English)
Graph Neural Networks (GNNs) are vital for learning from graph-structured data, enabling applications in network analysis, recommendation systems, and speech analytics. Deploying them on edge devices like client PCs and laptops enhances real-time processing, privacy, and cloud independence. GNNs aid Retrieval-Augmented Generation (RAG) for Large Language Models (LLMs) and enable event-based vision tasks. However, irregular memory access, sparsity, and dynamic structures cause high latency and energy overhead on resource-constrained devices. While modern edge processors integrate CPUs, GPUs, and NPUs, NPUs designed for data-parallel tasks struggle with irregular GNN computations. We introduce GraNNite, the first hardware-aware framework optimizing GNN execution on commercial-off-the-shelf (COTS) SOTA DNN accelerators via a structured three-step methodology: (1) enabling NPU execution, (2) optimizing performance, and (3) trading accuracy for efficiency gains. Step 1 employs GraphSplit for workload distribution and StaGr for static aggregation, while GrAd and NodePad handle dynamic graphs. Step 2 boosts performance using EffOp for control-heavy tasks and GraSp for sparsity exploitation. Graph Convolution optimizations PreG, SymG, and CacheG reduce redundancy and memory transfers. Step 3 balances quality versus efficiency, where QuantGr applies INT8 quantization, and GrAx1, GrAx2, and GrAx3 accelerate attention, broadcast-add, and SAGE-max aggregation. On Intel Core Ultra AI PCs, GraNNite achieves 2.6X to 7.6X speedups over default NPU mappings and up to 8.6X energy gains over CPUs and GPUs, delivering 10.8X and 6.7X higher performance than CPUs and GPUs, respectively, across GNN models.
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。