Dynamical Self-Reconfigurable Mechanism for Data-Driven Cell Array

doi:10.1007/s12204-021-2319-z

摘要/Abstract

摘要： The utilization of computation resources and reconfiguration time has a large impact on reconfiguration system performance. In order to promote the performance, a dynamical self-reconfigurable mechanism for datadriven cell array is proposed. Cells can be fired only when the needed data arrives, and cell array can be worked on two modes: fixed execution and reconfiguration. On reconfiguration mode, cell function and data flow direction are changed automatically at run time according to contexts. Simultaneously using an H-tree interconnection network, through pre-storing multiple application mapping contexts in reconfiguration buffer, multiple applications can execute concurrently and context switching time is the minimal. For verifying system performance, some algorithms are selected for mapping onto the proposed structure, and the amount of configuration contexts and execution time are recorded for statistical analysis. The results show that the proposed self-reconfigurable mechanism can reduce the number of contexts efficiently, and has a low computing time.

关键词: cell array , configurable computing , self-reconfigurable mechanism , data-driven , data flow graph

Abstract: The utilization of computation resources and reconfiguration time has a large impact on reconfiguration system performance. In order to promote the performance, a dynamical self-reconfigurable mechanism for datadriven cell array is proposed. Cells can be fired only when the needed data arrives, and cell array can be worked on two modes: fixed execution and reconfiguration. On reconfiguration mode, cell function and data flow direction are changed automatically at run time according to contexts. Simultaneously using an H-tree interconnection network, through pre-storing multiple application mapping contexts in reconfiguration buffer, multiple applications can execute concurrently and context switching time is the minimal. For verifying system performance, some algorithms are selected for mapping onto the proposed structure, and the amount of configuration contexts and execution time are recorded for statistical analysis. The results show that the proposed self-reconfigurable mechanism can reduce the number of contexts efficiently, and has a low computing time.

Key words: cell array , configurable computing , self-reconfigurable mechanism , data-driven , data flow graph

中图分类号:

SHAN Rui (山蕊), JIANG Lin (蒋林), WU Haoyue (吴昊玥), HE Feilong (贺飞龙), LIU Xinchuang (刘新闯). Dynamical Self-Reconfigurable Mechanism for Data-Driven Cell Array[J]. J Shanghai Jiaotong Univ Sci, 2021, 26(4): 511-521.

参考文献 30

[1]	YANG P L, MAREK-SADOWSKA M. Highperformance architecture using fast dynamic reconfigurable accelerators [J]. IEEE Transactions on Very Large Scale Integration (VLSI ) Systems, 2018, 26(7):1209-1222.
[2]	YIN S Y, YAO X Q, LU T Y, et al. Conflict-free loop mapping for coarse-grained reconfigurable architecture with multi-bank memory [J]. IEEE Transactions on Parallel and Distributed Systems, 2017, 28(9): 2471-2485.
[3]	TESSIER R, POCEK K, DEHON A. Reconfigurable computing architectures [J]. Proceedings of the IEEE,2015, 103(3): 332-354.
[4]	OCHI H, YAMAGUCHI K, FUJIMOTO T, et al. Viaswitch FPGA: Highly dense mixed-grained reconfigurable architecture with overlay via-switch crossbars[J]. IEEE Transactions on Very Large Scale Integration (VLSI ) Systems, 2018, 26(12): 2723-2736.
[5]	FAN X, WU D, CAOW, et al. Stream processing dualtrack CGRA for object inference [J]. IEEE Transactions on Very Large Scale Integration (VLSI ) Systems,2018, 26(6): 1098-1111.
[6]	DUCH L, BASU S, BRAOJOS R, et al. HEALWEAR:An ultra-low power heterogeneous system for bio-signal analysis [J]. IEEE Transactions on Circuits and Systems I : Regular Papers, 2017, 64(9): 2448-2461.
[7]	YIN S Y, YAO X Q, LIU D J, et al. Memory-aware loop mapping on coarse-grained reconfigurable architectures [J]. IEEE Transactions on Very Large Scale Integration (VLSI ) Systems, 2016, 24(5): 1895-1908.
[8]	KARUNARATNE M, MOHITE A K, MITRA T, et al. HyCUBE: A CGRA with reconfigurable single-cycle multi-hop interconnect [C]//Design Automation Conference.Austin: IEEE, 2017: 1-6.
[9]	JAFRI S MA H, GIA T N, DYTCKOV S, et al. NeuroCGRA:A CGRA with support for neural networks[C]// International Conference on High Performance Computing & Simulation. Bologna: IEEE, 2014: 506-511.
[10]	WANG Y S, LIU L B, YIN S Y, et al. 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 Transactions on Very Large Scale Integration (VLSI )Systems, 2014, 22(5): 983-994.
[11]	FELL A, R′AKOSSY Z E, CHATTOPADHYAY A. Force-directed scheduling for data flow graph mapping on coarse-grained reconfigurable architectures[C]//International Conference on ReConFigurable Computing and FPGAs. Cancun: IEEE, 2014:1-8.
[12]	MUKHERJEE M, FELL A, GUHA A. DFGenTool: A dataflow graph generation tool for coarse grain reconfigurable architectures [C]//2017 30th International Conference on VLSI Design and 2017 16th International Conference on Embedded Systems (VLSID). Hyderabad:IEEE, 2017: 67-72.
[13]	KRISHNAMOORTHY R, FUJITA M, VARADARAJAN K, et al. Interconnect-topology independent mapping algorithm for a Coarse Grained Reconfigurable Architecture [C]//International Conference on Field-Programmable Technology. New Delhi: IEEE, 2011: 1-5.
[14]	MATSUTANI H, KOIBUCHI M, YAMADA Y, et al. Fat H-tree: A cost-efficient tree-based on-chip network [J]. IEEE Transactions on Parallel and Distributed Systems, 2009, 20(8): 1126-1141.
[15]	LIU L B, WANG Y S, YIN S Y, et al. Row-based configuration mechanism for a 2-D processing element array in coarse-grained reconfigurable architecture [J].Science China Information Sciences, 2014, 57(10): 1-18.
[16]	WU L H, ZHANG X Z, PAN X L, et al. Design of an static reconfiguration based on FPGA system [J]. International Journal of Multimedia and Ubiquitous Engineering,2016, 11(2): 99-106.
[17]	JUNG S, KIM T G. An operation and interconnection sharing algorithm for reconfiguration overhead reduction using static partial reconfiguration [J]. IEEE Transactions on Very Large Scale Integration (VLSI )Systems, 2008, 16(12): 1589-1595.
[18]	LIU B, CAO P, ZHU M, et al. Reconfiguration process optimization of dynamically coarse grain reconfigurable architecture for multimedia applications [J]. IEICE Transactions on Information and Systems, 2012,E95-D(7): 1858-1871.
[19]	CAROTENUTO P L, DELLA CIOPPA A, MARCELLI A, et al. An evolutionary approach to the dynamical reconfiguration of photovoltaic fields [J]. Neurocomputing,2015, 170: 393-405.
[20]	GENG T S, LIU L B, YIN S Y, et al. Parallelization of computing-intensive tasks of the H.264 high profile decoding algorithm on a reconfigurable multimedia system [J]. IEICE Transactions on Information and Systems, 2010, E93-D(12): 3223-3231.
[21]	JAFRI S M A H, DANESHTALAB M, ABBAS N, et al. TransMap: Transformation based remapping and parallelism for high utilization and energy efficiency in CGRAs [J]. IEEE Transactions on Computers, 2016,65(11): 3456-3469.
[22]	KIM Y, JOO H, YOON S. Inter-coarse-grained reconfigurable architecture reconfiguration technique for efficient pipelining of kernel-stream on coarse-grained reconfigurable architecture-based multi-core architecture[J]. IET Circuits, Devices & Systems, 2016, 10(4): 251-265.
[23]	MOGHADDAMM S, BALAKRISHNANM, PAUL K.Partial reconfiguration for dynamic mapping of task graphs onto 2D mesh platform [M]//Applied reconfigurable computing. Cham: Springer International Publishing,2015: 373-382.
[24]	JAFRI S M A H, HEMANI A, PAUL K, et al. Compression based efficient and agile configuration mechanism for coarse grained reconfigurable architectures[C]//2011 IEEE International Symposium on Parallel and Distributed Processing Workshops and Phd Forum.Anchorage: IEEE, 2011: 290-293.
[25]	XIE Z, LU Y H, FAN Y B, et al. Data mapping scheme and implementation for high-throughput DCT/IDCT transpose memory [C]//IEEE International Conference on Solid-State and Integrated Circuit Technology.Guilin: IEEE, 2014: 1-3.
[26]	LIU W, LIU L B, YIN S Y, et al. A high parallel motion compensation implementation on a coarse-grained reconfigurable processor supporting H.264 high profile decoding [C]//IEEE International Conference on Solid-State and Integrated Circuit Technology. Guilin:IEEE, 2014: 1-3.
[27]	SANGHVI H. 2D cache architecture for motion compensation in a 4K Ultra-HD AVC and HEVC video codec system [C]//2014 IEEE International Conference on Consumer Electronics. Las Vegas: IEEE, 2014:189-190.
[28]	SHYMNA NIZAR N S, KRISHNA A R. An efficient folded pipelined architecture for Fast Fourier Transform using Cordic algorithm [C]//2014 IEEE International Conference on Advanced Communications, Control and Computing Technologies. Ramanathapuram:IEEE, 2014: 462-467.
[29]	ZHU J F, LIU L B, YIN S Y, et al. A hybrid reconfigurable architecture and design methods aiming at control-intensive kernels [J]. IEEE Transactions on Very Large Scale Integration (VLSI ) Systems, 2015,23(9): 1700-1709.
[30]	LIU L B, WANG D, ZHU M, et al. An energyefficient coarse-grained reconfigurable processing unit for multiple-standard video decoding [J]. IEEE Transactions on Multimedia, 2015, 17(10): 1706-1720.