内存制约加速比模型及其对计算的影响

孙贤和, 鲁潇阳 - 计算机科学技术学报, 2023 - jcst.ict.ac.cn
… the rethinking of computer architecture and operating system … memory hierarchy, the memory
latency function shows a four… global performance optimization of the overall performance

基于制导的GPU 共享内存相关优化

李晶, 刘雷, 吴远, 刘向华, 高翊, 冯晓兵… - 计算机科学技术学报, 2016 - jcst.ict.ac.cn
optimizations eliminate bank and channel conflicts and expose parallel instructions, according
to underlying architecture… , since off-chip memory latency needs more parallel instructions …

在大规模系统上优化TPC-C 评测程序

J Zhai, F Zhang, Q Li, W Chen, W Zheng - Science China Information …, 2016 - Springer
… -level optimization methods to improve the TPC-C performance. … The main modules in the
LVM architecture are physical … The main bottleneck of this server was memory latency. Through …

多核集群上稀疏矩阵向量乘法的自动调优

SG Li, CJ Hu, JC Zhang, YQ Zhang - Science China Information Sciences, 2015 - Springer
… Our model-driven auto-tuning method forms a basis for tuning irregular applications on such
future architectures. In the future work, we plan to add optimization parameters for a single …

一种消息式内存系统

陈荔城, 陈明宇, 阮元, 黄永兵, 崔泽汉, 卢天越… - 计算机科学技术 …, 2014 - jcst.ict.ac.cn
… on memory architecture as well as new optimization … contribute up to 70% of the total memory
latency in an 8-core system[1]. … They suggested two approaches: the first one is to distribute …

[PDF][PDF] Department of Electrjcal Engineering and Computer Science 巴May 11, 1988

RA Iannucci - 1988 - academia.edu
architectures that have directly reduced the effect of memory latency on performance are …
It is only in those situations where sequencing cannot be so optimized at compile time, eg, for …

面向资源解耦合架构的云计算系统软件

王晨曦, 单一舟, 左鹏飞, 崔慧敏 - 计算机科学技术学报, 2023 - jcst.ict.ac.cn
… disaggregated architecture brings severe performance and … specialized optimizations for
disaggregated architectures, … applications are more memory latency sensitive and perform …

基于保留收益的智能高速缓存替换

李凌达, 陆俊林, 程旭 - 计算机科学技术学报, 2014 - jcst.ict.ac.cn
… disparity in memory latency will become even larger in future … the performance of this method
is similar to that of the simple … candidate in computer architecture of Peking University. His …

[PDF][PDF] 巴团g 喹

S Owicki, A Agarwal - 1989 - bitsavers.org
… -processor caches to reduce memory latency and to avoid contention … While we focused on
a simple network architecture in this … For example, the compiler can optimize performance by …

共享缓存性能的度量与分析

丁晨, 向晓娅, 包斌, 罗昊, 罗英伟, 汪小林 - 计算机科学技术学报, 2014 - jcst.ict.ac.cn
performance across different computer architectures. Beyls and D’Hollander[30-31] built a
program tuning … linear correlation between the memory latency and the memory bandwidth …