arrow
返回

VMT: Virtualized Multi-Threading for Accelerating Graph Workloads on Commodity Processors

delete2022-06-01
delete1
delete
OA
AI
J
Josué Feliu *
A
Ajeya Naithani
J
Julio Sahuquillo
S
Salvador Petit
M
Moinuddin K. Qureshi
L
Lieven Eeckhout
DOI:10.1109/TC.2021.3086069delete
delete原文链接
delete分享
delete收藏
查看原文
摘要

摘要

En 中文
Modern-day graph workloads operate on huge graphs through pointer chasing which leads to high last-level cache (LLC) miss rates and limited memory-level parallelism (MLP). Simultaneous Multi-Threading (SMT) effectively hides the memory access latencies for multi-threaded graph workloads provided that sufficient threads are supported in hardware. Unfortunately, providing a sufficiently large number of physical threads incurs an unjustifiably high hardware cost for commodity SMT processors which typically implement only two physical hardware threads. Ideally, we would like to achieve aggressive-SMT performance when running graph workloads on modest commodity processors. In this paper, we propose Virtualized Multi-Threading (VMT), a low-overhead multi-threading paradigm for accelerating graph workloads on commodity processors. Unlike prior multi-threading paradigms, VMT virtualizes both the physical hardware threads and the architecture state: VMT maps a large number of logical software threads to a small number of physical hardware threads, while maintaining the architecture state of the logical threads in the processor's cache hierarchy. Implemented on top of a quad-core 2-way SMT processor, VMT achieves an average speedup of 1.74x for a set of representative graph workloads, while incurring minimal hardware cost (195 bytes per core to support up to 32 logical threads). VMT's low hardware cost paves the way for implementation in commodity processors.
Keyword:
Instruction sets
Computer architecture
Hardware
Registers
Software
Message systems
Switches
Architecture
multi-threading
virtualization
graph workloads
AI总结

AI总结

对已上传原文的论文进行重点信息的提取,主要内容包括:简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。

期刊

IEEE Transactions on Computers 封面图
IEEE Transactions on Computers
IF:
3.8
论文数:
5.3K
被引数:
9.8K

机构

U
Universitat Politecnica de Valencia
学者数:
1.5W
论文数: 1.4W
被引数: 18
G
Ghent University
学者数:
5.2W
论文数: 4.5W
被引数: 5.5W
U
University of Murcia
学者数:
9.2K
论文数: 8.1K
被引数: 8
U
university system of georgia
学者数:
7.3W
论文数: 6.5W
被引数: 101
学者 查看更多机构
引用论文

引用论文

err分享
err收藏
Imine-linked micron-network polymers with high polyethylene glycol uptake for shaped-stabilized phase change materials
err2016-01-01
err0
PREAI
errJia Tang; Shuang Fan; Wenjun Dong; Jingjing Wang; Hongyi Gao; Mu Yang; Ming Yang; Ge Wang
err分享
err收藏
Graph-based methods for analysing networks in cell biology
err2006-05-23
err374
errOAAI
errAittokallio, Tero; Schwikowski, Benno
err分享
err收藏
学者 查看更多内容