CAREER: Autotuning for multicore and manycore architectures: an enhanced feedback-driven approach

职业：多核和众核架构的自动调整：增强的反馈驱动方法

• 批准号: 1253292
• 负责人: Apan Qasem
• 金额: $ 48.95万
• 依托单位: Texas State University - San Marcos
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 2013
• 资助国家: 美国
• 起止时间: 2013-02-01 至 2020-01-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1253292&HistoricalAwards=false
• 关键词: CAREER; Autotuning; multicore; manycore; architectures

Achieving a high fraction of peak performance on complex architectures has been a perennial challenge for application developers. The emergence of multicore processors and accelerators has greatly exacerbated this problem. With an increasing number of cores per socket, deep hierarchies of shared and distributed caches, and exascale computing on the horizon, multicore platforms pose unprecedented challenges for software development and application tuning. This research confronts the challenge of multicore and manycore software development by improving automatic performance tuning through improved feedback diagnostics. Autotuning efficiency is achieved through enhanced knowledge of the problem domain, program features and architectural characteristics. To this end, we are developing a set of tools that allow specification, collection and synthesis of tuning related information. This rich set of information is then exploited through novel machine learning models to deliver scalable, portable and sustainable performance on diverse architectures. In line with this research, we have established, SISTEM, a structured interdisciplinary program for undergraduates that brings together faculty and students from different STEM disciplines to explore cross-cutting research problems in which computational thinking, modeling and simulation play a central role. The techniques developed as past of this research will have a direct economic impact through saved energy and computation cycles on high-end systems. The improved efficiency of applications will allow researchers to execute large-scale simulations that will enable modeling of complex phenomenon in a broad range of disciplines including medicine, high-energy physics, climate modeling and nanotechnology.

在复杂的体系结构上实现高比例的峰值性能一直是应用程序开发人员面临的长期挑战。多核处理器和加速器的出现大大加剧了这一问题。随着每个插槽的内核数量不断增加，共享和分布式缓存的深层层次结构，以及即将出现的亿级计算，多核平台对软件开发和应用程序调优提出了前所未有的挑战。这项研究通过改进反馈诊断来改进自动性能调整，从而面对多核和多核软件开发的挑战。自动调整效率是通过增强对问题域、程序功能和体系结构特征的了解来实现的。为此，我们正在开发一套工具，允许规范、收集和综合与调整相关的信息。然后，通过新的机器学习模型利用这些丰富的信息集，在不同的架构上提供可扩展、可移植和可持续的性能。根据这项研究，我们建立了一个面向本科生的结构化跨学科项目--SISTEM，它将来自不同STEM学科的教职员工和学生聚集在一起，探索以计算思维、建模和模拟为核心的跨学科研究问题。本研究过去开发的技术将通过在高端系统上节省能源和计算周期而产生直接的经济影响。应用程序效率的提高将使研究人员能够执行大规模模拟，从而能够在包括医学、高能物理、气候建模和纳米技术在内的广泛学科中对复杂现象进行建模。

项目成果

Energy-Efficient GPU Graph Processing with On-Demand Page Migration

具有按需页面迁移功能的节能 GPU 图形处理

• DOI: 10.1109/igsc48788.2019.8957183
• 发表时间: 2019
• 期刊: 2019 Tenth International Green and Sustainable Computing Conference (IGSC
• 作者: Hope, Jacob M;Nag, Trisha;Qasem, Apan
• 通讯作者: Qasem, Apan

Accelerating HotSpots in Deep Neural Networks on a CAPI-Based FPGA

• DOI: 10.1109/hpcc/smartcity/dss.2019.00048
• 发表时间: 2019-08
• 期刊: 2019 IEEE 21st International Conference on High Performance Computing and Communications; IEEE 17th International Conference on Smart City; IEEE 5th International Conference on Data Science and Systems (HPCC/SmartCity/DSS)
• 作者: Md. Syadus Sefat;S. Aslan;J. W. Kellington;Apan Qasem

Evaluating the impact of data layout and placement on the energy efficiency of heterogeneous applications

评估数据布局和放置对异构应用程序能效的影响

• DOI: 10.1109/igcc.2017.8323574
• 发表时间: 2017
• 期刊: 2017 Eighth International Green and Sustainable Computing Conference (IGSC
• 作者: Qasem, Apan;Teich, Samuel
• 通讯作者: Teich, Samuel

Automatically Selecting Profitable Thread Block Sizes for Accelerated Kernels

自动为加速内核选择有利可图的线程块大小

• DOI: 10.1109/hpcc-smartcity-dss.2017.58
• 发表时间: 2017
• 期刊: IEEE 19th International Conference on High Performance Computing and Communications;
• 作者: Connors, Tiffany A.;Qasem, Apan
• 通讯作者: Qasem, Apan

A Machine Learning Approach to Automatic Creation of Architecture-Sensitive Performance Heuristics

自动创建架构敏感的性能启发式的机器学习方法

• DOI: 10.1109/hpcc-smartcity-dss.2017.3
• 发表时间: 2017
• 期刊: IEEE 19th International Conference on High Performance Computing and Communications;
• 作者: Saha, Biplab Kumar;Connors, Tiffany A.;Rahman, Saami;Qasem, Apan
• 通讯作者: Qasem, Apan