|
|
论文:2012,Vol:30,Issue(6):961-967 |
|
|
引用本文： |
|
|
张萌, 赵磊, 樊晓桠, 田杭沛. 面向混杂流计算的适应性存储器体系结构[J]. 西北工业大学 |
|
|
Zhang Meng, Zhao Lei, Fan Xiaoya, Tian Hangpei. Exploring Hybrid Stream Computing Oriented On-Chip Adaptive Memory Structure[J]. Northwestern polytechnical university |
|
|
|
|
|
|
|
| 面向混杂流计算的适应性存储器体系结构 |
|
| 张萌, 赵磊, 樊晓桠, 田杭沛 |
|
| 西北工业大学 计算机学院, 陕西 西安 710072 |
| 摘要: |
| 可将科学计算中大量算法的计算形式视为由流计算和相当比例的通用计算混合而成。针对低并行度计算以及不易流化(Streamlization)的数据结构对流计算整体性能具有较大影响,提出了一种软、硬件可控的适应性片上存储结构DAMS Cache。该结构能够同时适应混杂流计算中流数据以及标量数据的存储需求;采用了适应性动态存储资源分配策略和适应性动态地址映射策略解决地址映射冲突问题;通过全硬件支持非规则流、条件流的存储与访问,混合数据替换策略能够充分挖掘数据的生产者-消费者局部性及时间、空间局部性。验证评估实验表明,相对Cache以及SPM(Scratchpad Memory),DAMS Cache算法的适应性较好,面向混杂流计算的性能较优。 |
| 关键词:
片内高速缓存
计算机体系结构
计算机硬件
计算机仿真
计算机软件
数据处理
高效率
微处理器芯片
多处理系统
优化
资源配置
调度
结构框图
DAMS动态地址映射流
|
|
| Exploring Hybrid Stream Computing Oriented On-Chip Adaptive Memory Structure |
|
| Zhang Meng, Zhao Lei, Fan Xiaoya, Tian Hangpei |
|
| Department of Automatic Control,Northwestern Polytechnical University,Xi'an 710072,China |
| Abstract: |
| In scientific applications,the computing process is composed of stream computing and general compu-ting. The performance of computing is limited by low parallelism and hard streamlization. We propose what we callDAMS-Cache (DAMS stands for Dynamical Address Mapping Stream),which can be controlled by both the soft-ware and hardware that manage an on-chip memory structure that can,we believe,suppress effectively the above-mentioned limitation. Sections 1 through 4 of the full paper explain DAMS-Cache,whose core consists of: (1) thememory system can support both stream computing and scalar data processing in high performance,which supportsirregular stream and conditionally-loaded stream computing by hardware; (2) the DAMS Cache explores bothcoarse-grained producer-consumer locality and fine-grained temporal/spatial locality by using replacement strategyof mixed data structure; (3) in order to avoid address mapping conflict,the dynamically adaptive storage resourceassignment strategy and dynamically adaptive address mapping strategy are also presented. Section 5 presents per-formance analysis of DAMS Cache; it shows preliminarily that DAMS Cache has better adaptability and high per-formance compared with Scratchpad Memory and traditional Cache. |
| Key words:
cache memory
computer architecture
computer hardware
computer simulation
computer software
data processing
efficiency
microprocessor chips
multiprocessing systems
optimization
resourceallocation
scheduling
schematic diagrams;DAMS (dynamical address mapping stream)
|
|
| 收稿日期: 2011-12-04
修回日期:
|
| DOI: |
| 基金项目: 高等学校博士学科点专项科研基金(20116102120049);西北工业大学基础研究基金(JC20120239)资助 |
| 通讯作者:
Email： |
| 作者简介: 张萌(1978-)，西北工业大学讲师、博士，主要从事多核处理器系统结构、处理器性能仿真模型研究。
|
|
| 相关功能 |
|
|
|
|
| 作者相关文章 |
|
| 张萌 在本刊中的所有文章 |
| 赵磊 在本刊中的所有文章 |
| 樊晓桠 在本刊中的所有文章 |
| 田杭沛 在本刊中的所有文章 |
|
|
|
|
|
|
|
|
参考文献: |
|
|
[1] Leverich J，Arakida H，Solomatnikov A，et al.Comparing Memory Systems for Chip Multiprocessors.Proceedings of the 34thAnnual International Symposium on Computer Architecture (ISCA'07)，San Diego，California，USA，ACM.2007, 358-368
[2] 田杭沛.面向混杂流计算的适应性体系结构研究:[博士学位论文].西安: 西北工业大学, 2010Tian H P.Research on Adaptive Architecture of Hybrid Stream Computing.Ph.D.Thesis，Xi'an，Northwestern PolytechnicalUniversity, 2010 (in Chinese)
[3] Rixner S，Dally W J，Kapasi U J，et al.A Bandwidth-Efficient Architecture for Media Processing.Proceedings of the 31st An-nual ACM/IEEE International Symposium on Microarchitecture (MICRO'98)，Dallas，Texas，United States, 1998
[4] Wen M, Wu N, Zhang C, et al.On-Chip Memory System Optimization Design for the FT64 Scientific Stream Accelerator.IEEEMicro, 2008, 28(4): 51-70
[5] Jayasena N S.Memory Hierarchy Design for Stream Computing.Ph.D.Thesis，California，United States，Stanford University．2005
[6] Merrimac M Erez.High-Performance and Highly-Efficient Scientific Computing with Streams.Ph.D.Thesis，California，UnitedStates，Stanford University, 2007 |
|
|
|
相关文献: |
|
|
1．郑曦, 智永锋, 张骏.伪仿射投影算法优化迭代步长的研究[J]. 西北工业大学, 2014,32(1): 102-105 |
|
|
2．智永锋, 郑曦, 李茹, 邓正宏, 张骏.一种变步长AP-REE算法的研究[J]. 西北工业大学, 2013,31(2): 183-188 |
|
|
3．苏进展, 方宗德, 蔡香伟.弧齿锥齿轮数字化齿面啮合仿真分析与试验[J]. 西北工业大学, 2012,30(4): 565-569 |
|
|
4．张奕, 程小辉, 蔡皖东, 詹涛.混合关键度驱动的实时调度研究[J]. 西北工业大学, 2012,30(4): 617-621 |
|
|
5．夏平, 周兴社, 谢必昌.适应动态安全需求的实时任务调度算法研究[J]. 西北工业大学, 2012,30(5): 657-662 |
|
|
6．田薇, 孙秦, 范学领.某型飞机前起舱段结构几何刚度分布优化设计[J]. 西北工业大学, 2012,30(6): 857-861 |
|
|
|
|
|
|
