Architecture and Evaluation of a Third-Generation RHiNET Switch for High-Performance Parallel Computing(<Special Issue> Development of Advanced Computer Systems)
スポンサーリンク
概要
- 論文の詳細を見る
RHiNET-3/SW is the third-generation switch used in the RHiNET-3 system. It provides both low-latency processing and flexible connection due to its use of a credit-based flow-control mechanism, topology-free routing, and deadlock-free routing. The aggregate throughput of RHiNET-3/SW is 80Gbps, and the latency is 140ns. RHiNET-3/SW also provides a hop-by-hop retransmission mechanism. Simulation demonstrated that the effective throughput at a node in a 64-node torus RHiNET-3 system is equivalent to the effective throughput of a 64-bit 33-MHz PCI bus and that the performance of RHiNET-3/SW almost equals or exceeds the best performance of RHiNET-2/SW, the second-generation switch. Although credit-based flow control requires 26% more gates than rate-based flow control to manage the virtual channels (VCs), it requires less VC memory than rate-based flow control. Moreover, its use in a network system reduces latency and increases the maximum throughput compared to rate-based flow control.
- 社団法人電子情報通信学会の論文
- 2003-10-01
著者
-
AMANO Hideharu
Keio University
-
Amano H
Keio Univ. Yokohama‐shi Jpn
-
Amano Hideharu
Keio Univ.
-
Amano Hideharu
Amano Laboratory Department Of Computer Science Keio University
-
Nishimura Shinji
Central Research Laboratory Hitachi Ltd.
-
Kudoh T
National Institute Of Advanced Industrial Science And Technology (aist)
-
Kudoh Tomohiro
National Institute Of Advanced Industrial Science And Technology
-
NISHI Hiroaki
Keio University
-
NISHIMURA Shinji
Hitachi, Ltd., Central Research Lab.
-
HARASAWA Katsuyoshi
Network Solutions System Group, Hitachi Hybrid Network Co., Ltd.
-
Nishimura S
Central Research Laboratory Hitachi Ltd.
-
Nishi Hiroaki
Keio Univ. Yokohama‐shi Jpn
-
Nishi H
Starc (semiconductor Technol. Academic Res. Center) Yokohama‐shi Jpn
-
Harasawa Katsuyoshi
Network Solutions System Group Hitachi Hybrid Network Co. Ltd.
関連論文
- Code Compression with Split Echo Instructions
- A Routing Algorithm for Multihop WDM Ring
- Wavelength Division Multiple Access Ring : Virtual Topology on a Simple Ring Network
- Code Compression with Split Echo Instructions
- Message Transfer Algorithms on the Recursive Diagonal Torus
- Performance Evaluation of Hardware Multi-process Execution on the Dynamically Reconfigurable Processor
- A High-Speed, Highly-Reliable Network Switch for Parallel Computing System Using Optical Interconnection(Special Issue on Optical Interconnects/Optical Signal Processing)
- BS-4-41 A Demonstration of Management and Control of Resources in Large-scale Network Environments(BS-4. Network Design, Management and Control for Future Networked Systems)
- Cache-Based Network Processor Architecture : Evaluation with Real Network Traffic(Novel Device Architectures and System Integration Technologies)
- Automatic Pipeline Construction Focused on Similarity of Rate Law Functions for an FPGA-based Biochemical Simulator
- Architecture and Evaluation of a Third-Generation RHiNET Switch for High-Performance Parallel Computing( Development of Advanced Computer Systems)
- The RDT Router Chip: A Versatile Router for Supporting a Distributed Shared Memory
- Grid Network Service-Web Services Interface Version 2 Achieving Scalable Reservation of Network Resources Across Multiple Network Domains via Management Plane
- A 100-Gb/s-Physical-Layer Architecture for Higher-Speed Ethernet for VSR and Backplane Applications(VLSI Architecture for Communication/Server Systems,VLSI Technology toward Frontiers of New Market)
- 100-Gb/s Physical-Layer Architecture for Next-Generation Ethernet( The Next Generation Ethernet Technologies)
- Signal Transmission and Coding Architecture for Next-Generation Ethernet(New Technologies in the Internet and their Applications)
- Self-Organized Link State Aware Routing for Multiple Mobile Agents in Wireless Network
- Low-Power Network-Packet-Processing Architecture Using Process-Learning Cache for High-End Backbone Router(Digital, Low-Power LSI and Low-Power IP)
- A Controller Design Method of Decentralized Control System
- An FPGA-Based Acceleration Method for Metabolic Simulation(Recornfigurable Systems)(Reconfigurable Systems)
- A Retargetable Compiler Based on Graph Representation for Dynamically Reconfigurable Processor Arrays
- Design and Implementation of RHiNET-2/NI0 : A Reconfigurable Network Interface for Cluster Computing
- Network Services Interface : An open standard for dynamic circuit service interoperability
- Performance Evaluation of 3-Dimensional MIN with Cache Consistency Maintenance Mechanism
- Design and Implementation Fine-grained Power Gating on Microprocessor Functional Units