SCIMA-SMP: on-chip memory processor architecture for SMP

Authors:
Chikafumi Takahashi;Masaaki Kondo;Taisuke Boku;Daisuke Takahashi;Hiroshi Nakamura;Mitsuhisa Sato
Affiliations:
University of Tsukuba;University of Tokyo;University of Tsukuba;University of Tsukuba;University of Tokyo;University of Tsukuba
Venue:
WMPI '04 Proceedings of the 3rd workshop on Memory performance issues: in conjunction with the 31st international symposium on computer architecture
Year:
2004

Citing 12
Cited 0

The cache performance and optimizations of blocked algorithms

ASPLOS IV Proceedings of the fourth international conference on Architectural support for programming languages and operating systems
BI-CGSTAB: a fast and smoothly converging variant of BI-CG for the solution of nonsymmetric linear systems

SIAM Journal on Scientific and Statistical Computing
Adapting cache line size to application behavior

ICS '99 Proceedings of the 13th international conference on Supercomputing
Performance of lattice QCD programs on CP-PACS

Parallel Computing - Special issue on high performance computing in lattice QCD
Reconfigurable caches and their application to media processing

Proceedings of the 27th annual international symposium on Computer architecture
Tiling optimizations for 3D scientific computations

Proceedings of the 2000 ACM/IEEE conference on Supercomputing
The architecture of the DIVA processing-in-memory chip

ICS '02 Proceedings of the 16th international conference on Supercomputing
OpenMP: An Industry-Standard API for Shared-Memory Programming

IEEE Computational Science & Engineering
A Case for Intelligent RAM

IEEE Micro
SCIMA: Software Controlled Integrated Memory Architecture for High Performance Computing

ICCD '00 Proceedings of the 2000 IEEE International Conference on Computer Design: VLSI in Computers & Processors
Analysis of Shared Memory Misses and Reference Patterns

ICCD '00 Proceedings of the 2000 IEEE International Conference on Computer Design: VLSI in Computers & Processors
Architecture and Compiler Co-Optimization for High Performance Computing

IWIA '02 Proceedings of the International Workshop on Innovative Architecture for Future Generation High-Performance Processors and Systems (IWIA'02)

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, we propose a processor architecture with programmable on-chip memory for a high-performance SMP (symmetric multi-processor) node named SCIMA-SMP (Software Controlled Integrated Memory Architecture for SMP) with the intent of solving the performance gap problem between a processor and off-chip memory. With special instructions which enable the explicit data transfer between on-chip memory and off-chip memory, this architecture is able to control the data transfer timing and its granularity by the application program, and the SMP bus is utilized efficiently compared with traditional cache-only architecture. Through the performance evaluation based on clock-level simulation for various HPC applications, we confirmed that this architecture largely reduces the bus access cycle by avoiding redundant data transfer and controlling the granularity of the data movement between on-chip and off-chip memory.