Pipeline-Based Partition Exploration for Heterogeneous Multiprocessor Synthesis
スポンサーリンク
概要
- 論文の詳細を見る
To achieve an automated implementation for the application-specific heterogeneous multiprocessor systems-on-chip (MPSoC), partitioning and mapping the sequential programs onto multiple parallel processors is one of the most difficult challenges. However, the existing traditional parallelizing techniques cannot solve the MPSoC-related problems effectively, so designers are still required to manually extract the concurrency potentials in the program. To solve this bottleneck, an automated application partition technique is needed. However, completely automatic parallelism is ineffective, so it is promising to explore concurrency for certain practical special structures. To settle those issues, this paper proposes a template-based algorithm to automatically partition a special load-compute-store (LCS) loop structure. Since specific-instruction customization for the application specific instruction-set processors (ASIPs) has interactions with task partitioning, the proposed algorithm integrates the dynamic pipelining and ASIP techniques using an iterative improvement strategy: first, an initial pipelining scheme is generated to obtain the maximum parallelism; second, under the primary partition results specific instructions are customized respectively for each subprogram; third, the program is repartitioned via pipelining under the specific instruction configurations. The proposed method has been implemented in the context of a commercial extensible multiprocessor design flow, using the Xtensa-based XTMP platform from Tensilica Inc. Based on a case study of Fast Fourier Transform (FFT), the experimental results indicate that the partitioned programs by the proposed method demonstrate an average speedup of 10× compared to the original sequential programs which have not been partitioned and run on the uniprocessor system.
- （社）電子情報通信学会の論文
- 2009-09-01
著者
-
GOTO Satoshi
Graduate School of Information, Production and Systems, Waseda University
-
Goto Satoshi
Waseda Univ. Jpn
-
ZHAO Kang
EDA lab, Department of Computer Science and Technology, Tsinghua University
-
BIAN Jinian
EDA lab, Department of Computer Science and Technology, Tsinghua University
-
DONG Sheqin
EDA lab, Department of Computer Science and Technology, Tsinghua University
-
SONG Yang
Graduate School of Information, Production and Systems, Waseda University
-
Song Yang
Graduate School Of Ips Waseda University
-
Song Yang
Graduate School Of Information Production And Systems Waseda University
-
Goto Satoshi
Graduate School Of Information Production And System Waseda University
-
Bian Jinian
Eda Lab Department Of Computer Science And Technology Tsinghua University
-
Zhao Kang
Eda Lab Department Of Computer Science And Technology Tsinghua University
-
Dong Sheqin
Eda Lab Department Of Computer Science And Technology Tsinghua University
-
Zhao Kang
Eda Lab Dep. Of Computer Sci. And Technol. Tsinghua Univ.
-
Bian Jinian
Eda Lab Dep. Of Computer Sci. And Technol. Tsinghua Univ.
-
Goto Satoshi
Graduate School of Infomlation Production and Systems Waseda University
関連論文
- Permutation Network for Reconfigurable LDPC Decoder Based on Banyan Network
- A-4-35 Constant Quality One-Pass Rate Control Algorithm in H.264/AVC
- BS-12-19 A Fast Block Type Decision Algorithm for H.264/AVC Intra Prediction(BS-12. Network Planning, Control, and Management)
- D-11-50 A Novel Fast Block Type Decision Algorithm for Intra Prediction in H.264/AVC High Profile
- D-11-14 A Mode Reduction Based Fast Integer Motion Estimation Algorithm for HDTV
- A-16-2 A Fast Mode Decision Algorithm for H.264/AVC Intra Prediction
- A LOW-POWER AND PERFORMANCE-AWARE DVB-S2 LDPC DECODER WITH LAYERED SCHEDULING
- Region-of-Interest based Preprocessing for H.264/AVC Encoding
- A Macroblock Homogeneity Detection Method and its Application for Block Size Decision in H.264/AVC
- Fast Inter Mode Decision Algorithm Based on Residual Feature
- Turbo decoding based on LDPC message passing algorithm
- Pipeline-Based Partition Exploration for Heterogeneous Multiprocessor Synthesis
- Efficient Fully-Parallel LDPC Decoder Design with Improved Simplified Min-Sum Algorithms(VLSI Architecture for Communication/Server Systems,VLSI Technology toward Frontiers of New Market)
- Power-Efficient LDPC Decoder Architecture Based on Accelerated Message-Passing Schedule(VLSI Architecture,VLSI Design and CAD Algorithms)
- Partially-Parallel LDPC Decoder Achieving High-Efficiency Message-Passing Schedule(Selected Papers from the 18th Workshop on Circuits and Systems in Karuizawa)
- Reconfigurable Adaptive FEC System Based on Reed-Solomon Code with Interleaving(Adaptive Signal Processing, Recent Advances in Circuits and Systems-Part 1)
- BS-12-18 A Novel Hardware-Friendly Regular 3-Step Integer Motion Estimation Algorithm for H.264/AVC(BS-12. Network Planning, Control, and Management)
- A-16-11 Level C+ Bandwidth Reduction Method for MPEG-2 to H.264 Transcoding
- A-16-1 Group-Based Prediction Scheme on Multiple Reference Frame Fractional Motion Estimation in H.264/AVC
- A Sorting-based Architecture of Finding the First Two Minimum Values
- Lossless VLSI Oriented Full Computation Reusing Algorithm for H.264/AVC Fractional Motion Estimation(Selected Papers from the 19th Workshop on Circuits and Systems in Karuizawa)
- D-11-70 Adaptive Spatial EC based on Numerical Measures of Edge Statistical Model
- An adaptive error concealment order in H.264/AVC (VLSI設計技術)
- An adaptive error concealment order in H.264/AVC (集積回路)
- A-4-15 Cross Low Pass Filter Based Subsampling Algorithm for H.264/AVC Motion Estimation
- BS-12-20 VLSI Design of Level C Bandwidth Reduction Scheme for MPEG-2 to H.264/AVC Transcoding(BS-12. Network Planning, Control, and Management)
- Lossy Strict Multilevel Successive Elimination Algorithm for Fast Motion Estimation(Selected Papers from the 19th Workshop on Circuits and Systems in Karuizawa)
- Low-Power Partial Distortion Sorting Fast Motion Estimation Algorithms and VLSI Implementations(Advanced Image Technology)
- Motion content based search range prediction in variable block size motion estimation (第20回 回路とシステム軽井沢ワークショップ論文集) -- (映像応用)
- A Fine-Grain Scalable and Low Memory Cost Variable Block Size Motion Estimation Architecture for H.264/AVC(Integrated Electronics)
- A VLSI Architecture for Variable Block Size Motion Estimation in H.264/AVC with Low Cost Memory Organization(VLSI Architecture,VLSI Design and CAD Algorithms)
- Scalable VLSI Architecture for Variable Block Size Integer Motion Estimation in H.264/AVC(Selected Papers from the 18th Workshop on Circuits and Systems in Karuizawa)
- 261MHz Parallel Tree Architecture for Full Search Variable Block Size Motion Estimation in H.264/AVC
- 261MHz Parallel Tree Architecture for Full Search Variable Block Size Motion Estimation in H.264/AVC
- A strict successive elimination algorithm for fast motion estimation (映像信号処理)
- A VLSI Array Processing Oriented Fast Fourier Transform Algorithm and Hardware Implementation(VLSI Architecture, VLSI Design and CAD Algorithms)
- Geometrical, Physical and Text/Symbol Analysis Based Approach of Traffic Sign Detection System(Advanced Image Technology)
- A Highly Parallel Architecture for Deblocking Filter in H.264/AVC(Parallel and/or Distributed Processing Systems, Recent Advances in Circuits and Systems-Part1)
- Low Power LDPC Code Decoder Architecture Based on Intermediate Message Compression Technique
- A 41mW VGA【triple bond】fps Quadtree Video Encoder for Video Surveillance Systems
- Standard Deviation and Intra Prediction Mode Based Adaptive Spatial Error Concealment (SEC) in H.264/AVC
- Variable Block Size Motion Vector Retrieval Schemes for H.264 Inter Frame Error Concealment
- A high-parallelism reconfigurable permutation network for IEEE 802.11n/803.16e LDPC decoder (情報理論)
- An Irregular Search Window Reuse Scheme for MPEG-2 to H.264 Transcoding
- Pipeline-Based Partition Exploration for Heterogeneous Multiprocessor Synthesis
- Exploring Partitions Based on Search Space Smoothing for Heterogeneous Multiprocessor System
- Fast Custom Instruction Identification Algorithm Based on Basic Convex Pattern Model for Supporting ASIP Automated Design
- Permutation Network for Reconfigurable LDPC Decoder Based on Banyan Network
- Implementation of LDPC decoder for 802.16e (情報理論)
- An multi-rate LDPC decoder system on FPGA (情報理論)
- D-11-21 Low complexity decoding with frame-skipping for surveillance video
- An Efficient Motion Vector Coding Scheme Based on Prioritized Reference Decision
- A High Performance and Low Bandwidth Multi-Standard Motion Compensation Design for HD Video Decoder
- A 48 Cycles/MB H.264/AVC Deblocking Filter Architecture for Ultra High Definition Applications
- A High-Speed Design of Montgomery Multiplier
- Reconfigurable Variable Block Size Motion Estimation Architecture for Search Range Reduction Algorithm
- D-11-17 MOTION CORRELATION ADAPTION BASED INTER MODE DECISION ALGORITHM FOR H.264/AVC
- High Throughput VLSI Architecture of a Fast Mode Decision Algorithm for H.264/AVC Intra Encoding
- Adaptive Sub-Sampling Based Reconfigurable SAD Tree Architecture for HDTV Application
- Hardware-Oriented Early Detection Algorithms for 4×4 and 8×8 All-Zero Blocks in H.264
- VLSI Oriented Fast Motion Estimation Algorithm Based on Pixel Difference, Block Overlapping and Motion Feature Analysis
- A High Throughput LDPC Decoder Design Based on Novel Delta-value Message-passing Schedule
- A-4-32 Four Pattern Adaptive Down-sampling-Method for Variable Block Size Video Coding
- A Selective Video Encryption Scheme for MPEG Compression Standard(Application, Cryptography and Information Security)
- Adaptive Search Range Algorithms for Variable Block Size Motion Estimation in H.264/AVC
- Content-Aware Fast Motion Estimation for H.264/AVC
- Edge Block Detection and Motion Vector Information Based Fast VBSME Algorithm
- Parallel Improved HDTV720p Targeted Propagate Partial SAD Architecture for Variable Block Size Motion Estimation in H.264/AVC
- A-4-18 Integer Search Position Based Fast Motion Estimation in H.264/AVC
- Periodic Spectrum Transmission for Single-Carrier Transmission Frequency-Domain Equalization(Wireless Communication Technologies)
- A-5-34 Closed Form Expression for Bit Error Rate Performance of UWB Signal in the Presence of Inter-Symbol Interference
- SAD accumulation termination algorithm and early loop termination decision based high quality fast motion estimation for H.264/AVC (第21回 回路とシステム軽井沢ワークショップ論文集) -- (動画像符号化(2))
- Inter search mode reduction based parallel propagate partial SAD architecture for variable block size motion estimation in H.264/AVC (第20回 回路とシステム軽井沢ワークショップ論文集) -- (映像応用)
- A High Performance Partially-Parallel Irregular LDPC Decoder Based on Sum-Delta Message Passing Schedule
- A-4-33 High Throughput Rate-1/2 Partially-Parallel Irregular LDPC Decoder
- D-11-15 A Novel Three-Step Error Concealment Method for H.264/AVC
- A Bandwidth Optimized, 64 Cycles/MB Joint Parameter Decoder Architecture for Ultra High Definition H.264/AVC Applications
- An Ultra-Low Bandwidth Design Method for MPEG-2 to H.264/AVC Transcoding
- Constrained Stimulus Generation with Self-Adjusting Using Tabu Search with Memory
- Voltage and Level-Shifter Assignment Driven Floorplanning
- A-6-12 Prediction-based Center-bias Fast Fractional Motion Estimation Algorithm for H.264/AVC
- Accurate Human Detection by Appearance and Motion
- Histogram of Template for Pedestrian Detection
- Generic Permutation Network for QC-LDPC Decoder
- D-11-20 An Adaptive Reference Frame Compression Scheme for Video Decoding
- ROI based Computational Complexity Reduction Scheme for H.264/AVC Encoder
- Scale invariance based salient contour detection (「膨大化する画像データのための高度画像入出力技術」[画像電子]学会創立35周年記念論文特集号)
- An efficient decoding algorithm for ISDB-S2 (情報理論)
- A Sorting-based Architecture of Finding the First Two Minimum Values
- Multiple Region-of-Interest Based H.264 Encoder with a Detection Architecture in Macroblock Level Pipelining
- A 530Mpixels/s Intra Prediction Architecture for Ultra High Definition H.264/AVC Encoder
- Greedy Optimization Algorithm for the Power/Ground Network Design to Satisfy the Voltage Drop Constraint
- Cache Based Motion Compensation Architecture for Quad-HD H.264/AVC Video Decoder
- A 115mW 1Gbps QC-LDPC Decoder ASIC for WiMAX in 65nm CMOS
- A 98GMACs/W 32-Core Vector Processor in 65nm CMOS
- A 6.72-Gb/s 8 pJ/bit/iteration IEEE 802.15.3c LDPC Decoder Chip
- Greedy Algorithm for the On-Chip Decoupling Capacitance Optimization to Satisfy the Voltage Drop Constraint
- A 115mW 1Gbps QC-LDPC Decoder ASIC for WiMAX in 65nm CMOS
- Cluster Generation and Network Component Insertion for Topology Synthesis of Application-Specific Network-on-Chips
- Correlated Noise Reduction for Electromagnetic Analysis