arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2308.13626cs.DS
收藏

SAGE:一种面向可扩展高效稀疏广义矩阵-矩阵乘法的基于存储的方法

SAGE: A Storage-Based Approach for Scalable and Efficient Sparse Generalized Matrix-Matrix Multiplication

Myung-Hwan Jang, Yunyong Ko, Hyuck-Moo Gwon, Ikhyeon Jo, Yongjun Park, Sang-Wook Kim

AI总结:

针对单机SpGEMM可扩展性不足、分布式方案通信开销大的问题,提出基于存储的SAGE方法,搭配三层架构与三项优化策略,在可扩展性和效率上优于现有方法。

摘要:

稀疏广义矩阵-矩阵乘法(SpGEMM)是现实世界网络分析中的一项基础运算。随着现实世界网络规模不断扩大,基于单机的SpGEMM方法无法对超出主存容量的大规模网络执行SpGEMM(即可扩展性不足)。尽管基于分布式系统的方法可借助多台机器处理大规模SpGEMM,但该方法因需聚合多台机器的结果而存在严重的机器间通信开销(即效率低下)。为解决这一困境,本文提出了一种新颖的基于存储的SpGEMM方法SAGE,该方法将给定网络存储在存储设备(如SSD)中,并通过三层架构仅在处理需要时将网络的必要部分加载到主存中。此外,我们指出了三个可能降低SAGE整体性能的挑战,并提出三种有效策略加以应对:(1)基于分块的工作负载分配,用于平衡线程间的工作负载;(2)内存内部分聚合,用于减少不必要的存储-内存I/O生成量;(3)分布感知内存分配,用于防止主存中出现意外的缓冲区溢出。通过大量评估,我们验证了SAGE在可扩展性和效率方面优于现有SpGEMM方法。

英文摘要:

Sparse generalized matrix-matrix multiplication (SpGEMM) is a fundamental operation for real-world network analysis. With the increasing size of real-world networks, the single-machine-based SpGEMM approach cannot perform SpGEMM on large-scale networks, exceeding the size of main memory (i.e., not scalable). Although the distributed-system-based approach could handle large-scale SpGEMM based on multiple machines, it suffers from severe inter-machine communication overhead to aggregate results of multiple machines (i.e., not efficient). To address this dilemma, in this paper, we propose a novel storage-based SpGEMM approach (SAGE) that stores given networks in storage (e.g., SSD) and loads only the necessary parts of the networks into main memory when they are required for processing via a 3-layer architecture. Furthermore, we point out three challenges that could degrade the overall performance of SAGE and propose three effective strategies to address them: (1) block-based workload allocation for balancing workloads across threads, (2) in-memory partial aggregation for reducing the amount of unnecessarily generated storage-memory I/Os, and (3) distribution-aware memory allocation for preventing unexpected buffer overflows in main memory. Via extensive evaluation, we verify the superiority of SAGE over existing SpGEMM methods in terms of scalability and efficiency.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑