An Instruction Fetch Policy Based on Multiple Fetch Priorities for SMT Processors

SUN Cai-xia; ZHANG Min-xuan

您当前的位置：

首页 >

文章列表页 >

An Instruction Fetch Policy Based on Multiple Fetch Priorities for SMT Processors

更新时间：2025-07-16

- An Instruction Fetch Policy Based on Multiple Fetch Priorities for SMT Processors
- Acta Electronica Sinica Vol. 34, Issue 5, Pages: 790-795(2006)
- 作者机构：
  
  国防科学技术大学计算机学院,湖南,长沙,410073
- 作者简介：
- 基金信息：
- DOI：
  CLC： TP303
- Published：2006
- 稿件说明：
移动端阅览
SUN Cai-xia, ZHANG Min-xuan. An Instruction Fetch Policy Based on Multiple Fetch Priorities for SMT Processors[J]. Acta Electronica Sinica, 2006, 34(5): 790-795.
DOI：

SUN Cai-xia, ZHANG Min-xuan. An Instruction Fetch Policy Based on Multiple Fetch Priorities for SMT Processors[J]. Acta Electronica Sinica, 2006, 34(5): 790-795. DOI：

摘要

同时多线程(SMT

Simultaneous Multithreading)处理器中

同时运行的线程在共享资源的同时也在竞争资源.如果一个发生L2 cache失效的线程长时间占用共享资源

那么会导致其他线程运行速度减慢

甚至会因为缺少资源而停顿下来

从而降低了SMT处理器的总体性能.本文提出了一种基于多个取指优先级的同时多线程取指策略MFP(Multiple Fetch Priorities)

用于减少L2 cache失效给处理器性能带来的负面影响.模拟结果表明

无论使用IPC作为度量标准还是使用Hmean作为度量标准

对于所有类型的工作负载

尤其是存储器访问密集的工作负载

MFP都要优于现有的其他取指策略.此外

对于不同的取指策略

MFP表现出不同程度的提升.相对于PDG的提升最明显

平均IPC以及平均Hmean分别提高了19.2%和27.7%.

Abstract

In Simultaneous Multithreading (SMT) processors

co-scheduled threads share the processor's resources

but at the same time compete for them.A thread missing in L2 cache may occupy most of available resources for a long time

causing other threads run slower than they could or even stall because of lack of resources.As a result

the overall performance of SMT processors is degraded.In this paper

we propose a novel fetch policy called MFP (Multiple Fetch Priorities) to prevent the negative effects caused by L2 cache misses.Results show that our policy outperforms previously proposed fetch policies for all types of workloads

especially for memory bounded workloads

whether using IPC as a metric or using the harmonic mean as a metric.Results also tell that our policy shows different degrees of improvement over other fetch policies.The increment over PDG is greatest

reaching 19.2% in IPC and 27.7% in Hmean on average.

关键词

Keywords

references

Views

962

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Joint Optimization of Distributed Intelligent Content Caching and User Association for CF-mMIMO

Robust Resource Allocation Algorithm for Secure Communication in LEO-Based Satellite Communication Systems

Random User Characteristics-Oriented Joint UAV Deployment and Resource Allocation Algorithm

A Learning-Inspired Resource Allocation for Energy Harvesting-Powered Cognitive M2M Communications

Crosstalk Aware Resource Allocation Method Based on Lightpath Load Balancing and Neighborhood Matching

Related Author

LI Yun

ZHANG Cheng-yu

YAO Zhi-xiu

XIA Shi-chao

TAN Zhen

XU Yong-jun

LU Cheng-zhuang

DONG Yi-heng

Related Institution

No.15 Institute of China Electronics Technology Group Corporation

School of Computer and Technology, Chongqing University of Posts and Telecommunications

School of Communications and Information Engineering, Chongqing University of Posts and Telecommunications

State Key Laboratory of Millimeter Waves, Southeast University

Aerospace New Generation Communications Co., Ltd.

⁰