A Memory Access Technology of Heterogeneous Multi-Core System Based on Cell Broadband Engine Architecture

冯国富; 董小社; 丁彦飞; 王旭昊

您当前的位置：

首页 >

文章列表页 >

A Memory Access Technology of Heterogeneous Multi-Core System Based on Cell Broadband Engine Architecture

更新时间：2025-12-08

- A Memory Access Technology of Heterogeneous Multi-Core System Based on Cell Broadband Engine Architecture
- Vol. 43, Issue 2, Pages: 1-5(2009)
- 作者机构：
  
  西安交通大学计算机科学与技术系,西安,710049
- 作者简介：
- 基金信息：
- DOI：
  CLC： TP393
- Online First：10 February 2009，
  
  Published：2009
- 稿件说明：
移动端阅览
冯国富, 董小社, 丁彦飞, et al. A Memory Access Technology of Heterogeneous Multi-Core System Based on Cell Broadband Engine Architecture[J]. 2009, 43(2): 1-5.
DOI：

冯国富, 董小社, 丁彦飞, et al. A Memory Access Technology of Heterogeneous Multi-Core System Based on Cell Broadband Engine Architecture[J]. 2009, 43(2): 1-5. DOI：

摘要

针对Cell 宽带引擎架构(CBEA)多核高性能处理器要求软件显式地对分层存储结构进行管理

带来架构的可编程性及性能等问题

提出了一种基于CBEA的异构多核访存技术.将CBEA访存分为批量访存和按需访存; 通过合理部署数据缓冲区来减小批量访存计算中的片内访存开销

利用支持粗粒度访问的软件管理cache及数据预取来降低按需访存的片外访存开销; 以访存接口库的方式来改善软件的可编程性.实验结果表明

所提技术的访存接口库在批量访存方式下的性能比ALF和CellSs提高了30%～50%

按需访存中软件管理cache性能比CBE软件开发工具包提高了20%～30%

4路数据预取访存比单路缓存的性能提高约50%.

Abstract

A memory access technology is proposed to solve the problems of programming and performance caused by explicit management for multi-level memory hierarchies of Cell broadband engine architecture(CBEA). The access to main memory on CBEA is classified into two major types: bulk data transfer and on-demand data transfer. For bulk data transfer

the overhead of on-chip memory access is reduced by proper organization of data buffer. Meanwhile

a software-managed cache supporting coarse-grained access and data pre-fetch are adopted to enhance the performance of on-demand off-chip data transfer. In order to facilitate programming

a memory access library based on the proposed methods is implemented. Experimental results show that the performance of applications based on the proposed bulk data transfer technology is about 30%-50% higher than that of applications based on ALF and CellSs. For on-demand data transfer

the proposed software-managed cache performs about 20%-30% better than that of CBE SDK does

and the data prefetch technology based on four-buffer gets about 50% performance increase compared to single-buffer cache.

关键词

Keywords

references

GSCHWIND M. Chip multiprocessing and the cell broadband engine [C]∥Proceedings of ACM Computing Frontiers. New York, USA: ACM Press, 2006:1-8.

OHARA M,INOUE H,SOHDA Y, et al. MPI microtask for programming the cell broadband engine processor [J]. IBM Systems Journal, 2006, 45(1): 85-102.

BELLENS P, PEREZ J M, BADIA R M, et al. CellSs: a programming model for the cell BE architecture [C]∥Proceedings of the ACM/IEEE SC 2006 Conference on High Performance Networking and Computing. New York, USA: ACM Press,2006:86.

EICHENBERGER A E, O'BRIEN J K, O'BRIEN K M, et al. Using advanced compiler technology to exploit the performance of the cell broadband engine [trademark] architecture [J]. IBM Systems Journal, 2006, 45(1): 59-84.

VAN DER SPOEL D, LINDABL E, HESS B, et al. Gromacs: fast, flexible, and free [J]. Journal of Computational Chemistry, 2005,26(16):1701-1718.

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Effects of Expansion Angle Crater Film Hole on the Film Cooling Characteristics of a Double-Wall Structure

Bi-Level Multi-Obj ective Optimization of a Wind-Solar Coupled Hydrogen Production System Based on the Synergy of Hybrid Electrolyzer Arrays

Energy Efficiency Optimization of the Cryogenic Air Separation Process for Electronic-Grade Ultra Pure Nitrogen

Turbulent Kinetic Energy Dissipation Mechanism of Supercritical Carbon Dioxide Leakage Flow in a Scallop Bionic Seal Cavity

I/O Performance Modeling and Prediction Method for Large-Scale Parallel Applications Using Machine Learning

Related Author

No data

Related Institution

No data

⁰