收稿日期: 2015-02-02
修回日期: 2016-01-20
网络出版日期: 2016-11-30
基金资助
国家自然科学基金（No.61404028，No.61204023）；国家“863”高技术研究发展计划基金（No.2012AA012703）资助
Efficient Management Mechanism for Configuration Cache of Reconfigurable System for Multimedia Applications
Received date: 2015-02-02
Revised date: 2016-01-20
Online published: 2016-11-30
刘波, 龚宇, 季程, 王晓彤, 梅晨 . 一种面向媒体应用可重构系统的配置信息缓存管理机制[J]. 应用科学学报, 2016 , 34(6) : 724 -733 . DOI: 10.3969/j.issn.0255-8297.2016.06.008
The scheduling process of configuration context severely limits the performance of reconfigurable systems for multimedia applications. This paper proposes a management mechanism for configuration cache based on a hierarchical configuration context organization mechanism. It includes a configuration context prefetching mechanism based on preparsing and a hierarchical replacement strategy based on hybrid priorities. Experimental results show that the design described in this paper can improve configuration information access performance by 4.34 times. The proposed data cache structure has been applied in a reconfigurable system and implemented in a real chip. Processing performance of the reconfigurable system based on the proposed data cache structures is better than that of other reconfigurable architectures by more than 1.8 times.
[1] Kim Y, Mahapatra R N, Park I, Choi K. Low power reconfiguration technique for coarsegrained reconfigurable architecture [J]. IEEE Transactions on Very Large Scale Integration Systems, 2009, 17(5): 593-603.
[2] Wu B. B, Yan L, Wen Y, Chen T Z. Run-time configuration prefetching to reduce the overhead of dynamically reconfiguration [C]//IEEE International SOC Conference, Las Vegas, NV, USA, 2010: 305-308.
[3] Clemente J A, Resano J, Mozos D. An approach to manage reconfigurations and reduce area cost in hard real-time reconfigurable systems [J]. ACM Transactions on Embedded Computing Systems, 2014, 13(4): 1-24.
[4] Clemente J A, Perez Ramo E, Resano J. Configuration mapping algorithms to reduce energy and time reconfiguration overheads in reconfigurable systems [J]. IEEE Transactions on Very Large Scale Integration Systems, 2014, 22(6): 1248-1261.
[5] Bauer L, Shafique M, Henkel J. MinDeg: a performance-guided replacement policy for run-time reconfigurable accelerators [C]//7th IEEE/ACM International Conference on Hardware/ Software Codesign and System Synthesis, Grenoble, France, 2009: 335-342.
[6] Dybdahl H. An LRU-based replacement algorithm augmented with frequency of access in shared chip-multiprocessor caches [C]//Proceedings of the 2006 Workshop on Memory Performance: Dealing with Applications, Systems and Architectures, Seattle, Washington, 2006: 45-52.
[7] Wu C J, Jaleel A, Hasenplaugh W. SHiP: signature-based hit predictor for high performance caching [C]//Proceedings of the 44th Annual IEEE/ACM International Symposium on Microarchitecture. Porto Alegre, Brazil, 2011: 430-441.
[8] Kharbutli M, Yan S. Counter-based cache replacement and bypassing algorithms [J]. IEEE Transactions on Computers, 2008, 57(4): 433-447.
[9] Liu B, Cao P, Yang J J. Hybrid-priority configuration cache supervision method for coarse grained reconfigurable architecture [C]//2012 International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery, Sanya, China, 2012: 408-414.
[10] 刘波. 面向媒体处理可重构系统中存储子系统的关键技术研究[D]. 南京:东南大学,2013.
[11] 肖建. 面向人脸检测应用的可重构关键技术研究与实现[D]. 南京:东南大学,2014.
[12] Liu B, Cao P, Zhu M. Reconfiguration process optimization of dynamically coarse grain reconfigurable architecture for multimedia applications [J]. IEEE Transactions on Information and System, 2012, 95(7): 1858-1871.
[13] Liu L B, Wang D, Zhu M. An energy-efficient coarse-grained reconfigurable processing unit for multiple-standard video decoding [J]. IEEE Transcations on multimedia, 2015, 17(10): 1706-1720.
[14] Wang Y S, Liu L B, Yin S Y. On-chip memory hierarchy in one coarse-grained reconfigurable architecture to compress memory space and to reduce reconfiguration time and data-reference time [J]. IEEE Transcations on Very Large Scale Integration Systems, 2013, 22(5): 983-994.
[15] Liu L B, Chen Y J, Yin S Y. Implementation of multi-standard video decoding algorithms on a coarse-grained reconfigurable multimedia processor [C]//IEEE International Symposium on Circuits and Systems (ISCAS), Beijing, China, 2013: 897-900.
[16] Mei B, Vernalde S, Verkest D, Man H D, Lauwereins R. ADRES: an architecture with tightly coupled VLIW processor and coarse-grained reconfigurable matrix [C]//13rd International Conference In proceeding of Field Programmable Logic and Applications, Lisbon, Portugal, 2003: 61-70.
[17] Janakiraman N, Nirmalkumar P, Akram S M. Coarse grained ADRES based MIMO-OFDM transceiver with new Radix-25 pipeline FFT/IFFT processor [J]. Circuits, Systems and Signal Processing, 2014, 34(3): 1-23.
[18] De Sutter B, Raghavan P, Lambrechts A. Coarse-grained reconfigurable array architectures
[M]//Handbook of signal processing systems. Springer New York, 2013: 553-592.
[19] Voros N S, Hübner M, Becker J. MORPHEUS: a heterogeneous dynamically reconfigurable platform for designing highly complex embedded systems [J]. ACM Transactions on Embedded Computing Systems, 2013, 12(3): 70.
[20] Rossi D, Mucci C, Campi F. Application space exploration of a heterogeneous run-time configurable digital signal processor [J]. IEEE Transactions on Very Large Scale Integration Systems, 2012, 21(2): 193-205.
/
| 〈 |
|
〉 |
