中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
首页 机构 成果 学者
  
  1. CAS IR Grid
  2. 计算技术研究所
  3. 中国科学院计算技术研究所
  4. 中国科学院计算技术研究所期刊论文
  5. 英文
IrGEMM: An Input-Aware Tuning Framework for Irregular GEMM on ARM and X86 CPUs

文献类型：期刊论文

作者Wei, Cunyang1,2; Jia, Haipeng3; Zhang, Yunquan1,2; Yao, Jianyu1,2; Li, Chendi1,2; Cao, Wenxuan1,2
刊名IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS
出版日期2024-09-01
卷号35期号:9页码:1672-1689
关键词Kernel Libraries Computer architecture Tuning Layout Optimization Codes Batch GEMM code generation compact GEMM dynamic programming TSMM
ISSN号1045-9219
DOI10.1109/TPDS.2024.3432579
英文摘要The matrix multiplication algorithm is a fundamental numerical technique in linear algebra and plays a crucial role in many scientific computing applications. Despite the high performance of mainstream basic linear algebra libraries for large-scale dense matrix multiplications, they exhibit poor performance when applied to matrix multiplication with irregular input. This paper proposes an input-aware tuning framework that accounts for application scenarios and computer architectures to provide high-performance irregular matrix multiplication on ARMv8 and X86 CPUs. The framework comprises two stages: the install-time stage and the run-time stage. The install-time stage utilizes our proposed computational template to generate high-performance kernels for general data layout and SIMD-friendly data layout. The run-time stage utilizes a tiling algorithm suitable for irregular GEMM to select the optimal kernel and link as an execution plan. Additionally, load-balanced multi-threaded optimization algorithms are defined to exploit the multi-threading capability of modern processors. Experiments demonstrate that the proposed IrGEMM framework can achieve significant performance improvements for irregular GEMM on both ARMv8 and X86 CPUs compared to other mainstream BLAS libraries.
资助项目National Key Research and Development Program of China[2023YFB3001701] ; National Natural Science Foundation of China[62372432]
WOS研究方向Computer Science ; Engineering
语种英语
WOS记录号WOS:001284974600001
出版者IEEE COMPUTER SOC
源URL[http://119.78.100.204/handle/2XEOYT63/39663]  
专题中国科学院计算技术研究所期刊论文_英文
通讯作者Jia, Haipeng
作者单位1.Chinese Acad Sci, Inst Comp Technol, Beijing 100864, Peoples R China
2.Univ Chinese Acad Sci, Beijing 100049, Peoples R China
3.Chinese Acad Sci, Inst Comp Technol, State Key Lab Processors, Beijing 100864, Peoples R China
推荐引用方式
GB/T 7714		 Wei, Cunyang,Jia, Haipeng,Zhang, Yunquan,et al. IrGEMM: An Input-Aware Tuning Framework for Irregular GEMM on ARM and X86 CPUs[J]. IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS,2024,35(9):1672-1689.
APA Wei, Cunyang,Jia, Haipeng,Zhang, Yunquan,Yao, Jianyu,Li, Chendi,&Cao, Wenxuan.(2024).IrGEMM: An Input-Aware Tuning Framework for Irregular GEMM on ARM and X86 CPUs.IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS,35(9),1672-1689.
MLA Wei, Cunyang,et al."IrGEMM: An Input-Aware Tuning Framework for Irregular GEMM on ARM and X86 CPUs".IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS 35.9(2024):1672-1689.

入库方式： OAI收割

来源：计算技术研究所

浏览0
下载0
收藏0
其他版本

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。