.png){kind=icon}
返回
Optimizing the LINPACK Algorithm for Large-Scale PCIe-Based CPU-GPU Heterogeneous Systems
.png){kind=icon}
PRE
AI
DOI：10.1109/TPDS.2021.3067731.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
There is a widening gap between GPU and other components (CPU, PCIe bus and communication network) in heterogeneous parallel system. The gap forces us to orchestrate cooperative execution among these components much more carefully than ever before. By taking the LINPACK benchmark as a case study, this article proposes a fine-grained pipelining algorithm on large-scale CPU-GPU heterogeneous cluster systems. First, we build an algorithmic model that reveals a new approach to GPU-centric and fine-grained pipelining algorithm design. Then, we present four model-driven pipelining algorithms that incrementally squeeze bubbles in the pipeline so that it is occupied by more useful floating-point calculations. The algorithms are implemented on both the AMD and NVIDIA GPU platforms. The finally optimized LINPACK program achieves 107 PFlops on 25, 600 GPUs (70 percent floating-point efficiency). Several insights have been drawn to suggest tradeoff of algorithm design, programming support, and architecture design.
Keyword:
Pipeline processing
Graphics processing units
Computer architecture
Supercomputers
Clustering algorithms
Programming
Optimization
LINPACK algorithm
software pipeline
performance model
heterogeneous computing
cluster
AI总结
对已上传原文的论文进行重点信息的提取，主要内容包括：简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。
期刊
IF:
6
论文数:
5.2K
被引数:
1.1W
机构
引用论文
分享收藏Effects of Concurrent Use of an Ankle Support with a Laterally Wedged Insole for Medial Knee Osteoarthritis.png)
PM&R
PM&RIF0
PREAI
分享收藏
分享收藏Lack of Systematic Topographic Difference between Attention and Reasoning Beta CorrelatesPLoS ONE
OAAI
PLoS ONEIF0
OAAI分享收藏
查看更多内容