Performance Evaluation of Instruction Set Architecture of MBP-light : a distributed memory controller for a large scale multiprocessor
スポンサーリンク
概要
- 論文の詳細を見る
The instruction set architecture of MBP-light, a dedicated processor for the DSM (Distributed Shared Memory) management of JUMP-1 is analyzed with a real prototype. The Buffer-Register Architecture proposed for MBP-core improves performance with 5.64% in the home cluster and 6.27% in a remote cluster. It appears that the dominant operations in the DSM management program are handling packet queues assigned into the local cluster. Thus, common RISC instructions, especially load/store instructions, are frequently used. Separating instruction and data memory improves performance with 33%. The results suggest that another alternative which provides separate on-chip cache and instructions dedicated for packet queue management is advantageous.
- 一般社団法人情報処理学会の論文
- 2003-06-24
著者
-
Suzuki Noriaki
Department of Mathematics, Meijo University
-
AMANO Hideharu
Department of Information and Computer Science, Keio University
-
Suzuki Noriaki
Department Of Computer Science Graduate School Of Keio University
-
Amano Hideharu
Department Of Computer Science Graduate School Of Keio University
-
SUZUKI NORIAKI
Department of Chemical Engineering, Shizuoka University
関連論文
- Carleson inequalities on parabolic Bergman spaces
- Compct Toeplitz operators on parabolic Bergman spaces
- 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 case-control study on risk factors for scleroderma
- A link removal methodology for application-specific (リコンフィギャラブルシステム)
- A Parametric Study of Packet-Switched FPGA Overlay Networks
- Implementation of Data Driven Applications on a Multi-Context Reconfigurable Device
- Flexor tenosynovitis of the hands as an initial manifestation of systemic lupus erythematosus
- 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)
- 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)
- Special Section on Reconfigurable Systems
- High-Speed Fully-Adaptable CRC Accelerators
- A characterization of real entire functions by polynomial approximation for exponential weights (Potential Theory and its Related Fields)
- Nonintegrability of harmonic functions in a domain
- Schatten class Toeplitz operators on the parabolic Bergman space II