论文标题
阶段，模态，时间和空间位置：用于加速图分析的特定域特异性ML预摘要
Phases, Modalities, Temporal and Spatial Locality: Domain Specific ML Prefetcher for Accelerating Graph Analytics
论文作者
论文摘要
记忆性能是图形分析加速的瓶颈。现有的机器学习（ML）预摘要与相位过渡和图形处理中的不规则内存访问斗争。我们提出了MPGraph，这是一种基于ML的预摘要，用于使用域特定模型的图形分析。 MPGraph引入了三个新的优化：相变的软检测，用于访问三角洲和页面预测的相位特异性多模式模型，以及用于预取控制的链时时空预取（CSTP）。与Kolmogorov-Smirnov的窗口和决策树相比，我们的过渡探测器的精度高34.17-82.15％。与LSTM和Vanilla注意模型相比，我们的预测因素的F1得分达到6.80-16.02％的F1得分，PAGE预测的精度为11.68-15.41％。使用CSTP，MPGraph实现了12.52-21.23％的IPC改进，超过最先进的非ML预餐BO，降低了7.58-12.03％，基于ML的Prefetchers Voyager和Transfetch and Transfetch and Transfetch and Transfetch af 3.27-4.58％。对于实际实施，我们使用延迟降低的压缩模型证明了MPGraph，与BO相比，延迟的准确性和覆盖率明显高，IPC提高了3.58％。
Memory performance is a bottleneck in graph analytics acceleration. Existing Machine Learning (ML) prefetchers struggle with phase transitions and irregular memory accesses in graph processing. We propose MPGraph, an ML-based Prefetcher for Graph analytics using domain specific models. MPGraph introduces three novel optimizations: soft detection for phase transitions, phase-specific multi-modality models for access delta and page predictions, and chain spatio-temporal prefetching (CSTP) for prefetch control. Our transition detector achieves 34.17-82.15% higher precision compared with Kolmogorov-Smirnov Windowing and decision tree. Our predictors achieve 6.80-16.02% higher F1-score for delta and 11.68-15.41% higher accuracy-at-10 for page prediction compared with LSTM and vanilla attention models. Using CSTP, MPGraph achieves 12.52-21.23% IPC improvement, outperforming state-of-the-art non-ML prefetcher BO by 7.58-12.03% and ML-based prefetchers Voyager and TransFetch by 3.27-4.58%. For practical implementation, we demonstrate MPGraph using compressed models with reduced latency shows significantly superior accuracy and coverage compared with BO, leading to 3.58% higher IPC improvement.
