arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2408.15489cs.AR
收藏

Shared-PIM:为更快的DRAM内处理实现并发计算与数据流

Shared-PIM: Enabling Concurrent Computation and Data Flow for Faster Processing-in-DRAM

Ahmed Mamdouh, Haoran Geng, Michael Niemier, Xiaobo Sharon Hu, Dayane Reis

AI总结:

Shared-PIM通过在DRAM内存库中策略性分配行并利用内存外设,实现同一内存库内并发计算与数据传输,相比LISA降低数据移动延迟和能耗,集成到pLUTo后显著提升矩阵乘法、多项式乘法、数值传输及图处理任务性能,面积开销仅7.16%。

摘要:

内存内处理(PIM)赋予内存计算能力,有望解决内存与处理器之间数据传输相关的能耗和延迟问题。然而,在PIM架构中管理并发计算与数据流会为应用带来显著的延迟和能耗开销。本文提出Shared-PIM,这是一种面向DRAM内PIM的架构,通过内存外设的增强,在内存库中策略性地分配行,以支持并发处理和数据移动。Shared-PIM能够在同一内存库内同时进行计算和数据传输。与支持DRAM内PIM数据传输的最先进架构LISA相比,Shared-PIM将数据移动延迟和能耗分别降低了5倍和1.2倍。此外,当集成到最先进(SOTA)的DRAM内PIM架构(pLUTo)中时,Shared-PIM使加法和乘法速度提升1.4倍,从而将矩阵乘法(MM)任务性能提高40%,多项式乘法(PMM)提高44%,数值传输(NTT)任务提高31%。此外,对于广度优先搜索(BFS)和深度优先搜索(DFS)等图处理任务,Shared-PIM实现了29%的速度提升,而相对于基线pLUTo,面积开销仅为7.16%。

英文摘要:

Processing-in-Memory (PIM) enhances memory with computational capabilities, potentially solving energy and latency issues associated with data transfer between memory and processors. However, managing concurrent computation and data flow within the PIM architecture incurs significant latency and energy penalty for applications. This paper introduces Shared-PIM, an architecture for in-DRAM PIM that strategically allocates rows in memory banks, bolstered by memory peripherals, for concurrent processing and data movement. Shared-PIM enables simultaneous computation and data transfer within a memory bank. When compared to LISA, a state-of-the-art architecture that facilitates data transfers for in-DRAM PIM, Shared-PIM reduces data movement latency and energy by 5x and 1.2x respectively. Furthermore, when integrated to a state-of-the-art (SOTA) in-DRAM PIM architecture (pLUTo), Shared-PIM achieves 1.4x faster addition and multiplication, and thereby improves the performance of matrix multiplication (MM) tasks by 40%, polynomial multiplication (PMM) by 44%, and numeric number transfer (NTT) tasks by 31%. Moreover, for graph processing tasks like Breadth-First Search (BFS) and Depth-First Search (DFS), Shared-PIM achieves a 29% improvement in speed, all with an area overhead of just 7.16% compared to the baseline pLUTo.

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。

↑