Exploring PIM Architecture for High-Performance Graph Pattern Mining
Exploring PIM Architecture for High-Performance Graph Pattern Mining
复制标题
DOI：
10.1109/lca.2021.3103665
复制
发表时间：
2021-07
影响因子：
2.3
通讯作者：
Jiya Su;Linfeng He;Peng Jiang;Rujia Wang
中科院分区：
文献类型：
--
作者：
Jiya Su;Linfeng He;Peng Jiang;Rujia Wang
图挖掘应用程序（例如子图模式匹配和挖掘）广泛应用于生物信息学、社交网络分析和计算机视觉等现实领域。此类应用程序被认为是一类新型数据密集型应用程序，会产生大量不规则计算工作负载和内存访问，这与许多经过充分研究的图形应用程序（例如 BFS 和页排序）不同。在这封信中，我们使用新兴的内存处理架构来加速图挖掘任务中的数据密集型操作。我们首先确定最适合 PIM 执行的代码块。然后，我们观察到 PIM 架构上存在显着的负载不平衡，并分析了图挖掘应用程序中这种不平衡的根本原因。最后，我们评估了几种有助于减少负载不平衡的调度方案，并讨论了进一步提高性能的潜在优化。
Graph mining applications, such as subgraph pattern matching and mining, are widely used in real-world domains such as bioinformatics, social network analysis, and computer vision. Such applications are considered as a new class of data-intensive applications that generate massive irregular computation workloads and memory accesses, which are different from many well-studied graph applications such as BFS and page rank. In this letter, we use the emerging process-in-memory architecture to accelerate data-intensive operations in graph mining tasks. We first identify the code blocks that are best suitable for PIM execution. Then, we observe a significant load imbalance on PIM architecture and analyze the root cause for such imbalance in graph mining applications. Lastly, we evaluate several scheduling schemes that help reduce the load imbalance and discuss potential optimizations to enhance performance further.