A Parametric Study of Packet-Switched FPGA Overlay Networks
スポンサーリンク
概要
- 論文の詳細を見る
The constantly upgrading gate capacity of FPGAs enables us to implement a complex system on a chip. A packet-switched network presents to share network resources by multiple connections, so as to make the best use of link bandwidth. This study investigates the suitable overlay interconnects on FPGAs in terms of the amount of hardware, and throughput based on a parametric approach. Because the number of ports on a router sometimes dominates the amount of hardware for router, and its throughput performance, it has been chosen as a parameter. Based on a typical implementation of NoC router, various networks have been generalized and evaluated, in the case of both 16 hosts and 36 hosts. Evaluation results show that for small systems with 16 hosts or less, a full-crossbar switch is advantageous from the viewpoint of both throughput performance and hardware cost. On the other hand, when systems become large, the partitioned networks are efficient from the viewpoint of hardware cost. When the performance requirement is not so critical, we should select partitioned topology which requires minimum hardware and use some localization methods to improve the performance.
- 社団法人電子情報通信学会の論文
- 2006-09-08
著者
-
MATSUTANI Hiroki
Department of Information and Computer Science, Keio University
-
WANG Daihan
Department of Information and Computer Science, Keio University
-
KOIBUCHI Michihiro
National Institute of Informatics
-
AMANO Hideharu
Department of Information and Computer Science, Keio University
-
Amano Hideharu
Keio Univ. Yokohama Jpn
-
Matsutani Hiroki
Keio Univ. Yokohama Jpn
-
Matsutani Hiroki
Graduate School Of Science And Technology Keio University
-
Wang Daihan
Department Of Information And Computer Science Keio University
-
AMANO Hideharu
Faculty of Science and Technology, Keio University
-
YOSHIMI Masato
Department of Information and Computer Science, Keio University
-
Yoshimi Masato
Department Of Information And Computer Science Keio University
-
Amano Hideharu
Faculty Of Science And Technology Keio University
-
Amano Hideharu
Department Of Computer Science Keio University
-
Amano Hideharu
Department Of Computer Science Graduate School Of Keio University
関連論文
- Performance and cost evaluations of on-chip network topologies in FPGAs (リコンフィギャラブルシステム)
- A Leakage Efficient Data TLB Design for Embedded Processors
- Architectural Design of Next-Generation Science Information Network(Advanced Transfer Technologies for the Next Generation Network)
- Evaluation of a multicore reconfigurable architecture (VLSI設計技術)
- Evaluation of a multicore reconfigurable architecture (システムLSI設計技術)
- Evaluation of a multicore reconfigurable architecture (リコンフィギャラブルシステム)
- Evaluation of a multicore reconfigurable architecture (コンピュータシステム)
- An On/Off Link Regulation for Low-Power InfiniBand
- A Link Removal Methodology for Application-Specific Networks-on-Chip on FPGAs
- A link removal methodology for application-specific (リコンフィギャラブルシステム)
- A Port Combination Methodology for Application-Specific Networks-on-Chip on FPGAs(Reconfigurable System and Applications,Reconfigurable Systems)
- A temporal correlation based port combination methodology for application-specific networks-on-chip on FPGAs (計算機アーキテクチャ・2007年並列/分散/協調処理に関する『旭川』サマー・ワークショップ(SWoPP旭川2007)--研究会・連続同時開催)
- A Parametric Study of Packet-Switched FPGA Overlay Networks
- Implementation of Data Driven Applications on a Multi-Context Reconfigurable Device
- The MDX (Multi-Dimensional X'bar): A Class of Networks for Large Scale Multiprocessors (Special Issue on Architectures, Algorithms and Networks for Massively Parallel Computing)
- Performance Evaluation of Instruction Set Architecture of MBP-light : a distributed memory controller for a large scale multiprocessor
- A Dynamically Adaptive Hardware on Dynamically Reconfigurable Processor(Software Defined Radio Technology and Its Applications)
- A Leakage Efficient Data TLB Design for Embedded Processors
- Data Multicasting Procedure for Increasing Configuration Speed of Coarse Grain Reconfigurable Devices(Computer Systems)
- A Survey on Dynamically Reconfigurable Processors(Software Defined Radio Technology and Its Applications)
- Design and implementation of reconfigurable sensing system for networked robots
- Design and Implementation of RHiNET-2/NI0 : A Reconfigurable Network Interface for Cluster Computing
- Performance Evaluation of Instruction Set Architecture of MBP-Light in JUMP-1( Development of Advanced Computer Systems)
- Fault Tolerance of the TBSF (Tandem Banyan Switching Fabrics) and PBSF (Piled Banyan Switching Fabrics) (Special Issue on Architectures, Algorithms and Networks for Massively Parallel Computing)
- A Leakage Efficient Instruction TLB Design for Embedded Processors
- Special Section on Reconfigurable Systems
- High-Speed Fully-Adaptable CRC Accelerators
- Fine-Grained Run-Tume Power Gating through Co-optimization of Circuit, Architecture, and System Software Design
- A Fully Optical Ring Network-on-Chip with Static and Dynamic Wavelength Allocation