CAREER: An adaptive framework to accelerate real-time workloads in heterogeneous and reconfigurable environments
职业:一个自适应框架,可在异构和可重新配置的环境中加速实时工作负载
基本信息
- 批准号:2046444
- 负责人:
- 金额:$ 53.31万
- 依托单位:
- 依托单位国家:美国
- 项目类别:Continuing Grant
- 财政年份:2021
- 资助国家:美国
- 起止时间:2021-09-01 至 2026-08-31
- 项目状态:未结题
- 来源:
- 关键词:
项目摘要
Artificial intelligence and machine learning are enabling real-time decisions based on live data for interactive scientific discovery and mission critical applications such as autonomous driving and smart grid. They are increasingly powered by heterogeneous and even reconfigurable accelerators. The reconfigurability and heterogeneity of accelerators, together with stringent performance requirements and complex dependencies in real-time workloads, bring daunting operational challenges. These issues, if left unaddressed, would slow down scientific discovery and waste lots of computing resources and energy. This project will develop a heterogeneity and reconfigurability aware framework to accelerate real-time artificial intelligence and machine learning without hurting other workloads. It will benefit the society by improving the efficiency of costly computing systems, which saves taxpayers' money and better utilize existing investments. Real-time artificial intelligence and machine learning powered by the framework can better serve the society, e.g., accelerating scientific discovery and enabling data-driven control. The project will bring innovative education, outreach and training opportunities for both academic and industrial participants to train the next generation of researchers and practitioners for the society.Today, managing heterogeneous and reconfigurable systems for diverse workloads with high resource utilization and performance guarantee is an extremely challenging task. This project will design and implement an adaptive framework which automatically detects, profiles, and analyzes both workloads and accelerators on the fly. Based on the information, it adaptively reconfigures them to match resource capabilities with workload needs. Global and local optimization will be used to accommodate multiple types of workloads and the configuring, partitioning, placement, scheduling, and execution of models in each workload. The developed framework will provide provable performance even with partial information in unknown environments, which is urgently needed due to the ever increasing system complexity and volatility in workloads. Novel global resource allocation policies will be developed based on optimization techniques in this project to provide performance guarantee such as fairness, strategyproofness, and Pareto efficiency. Throughout the project, a reciprocal methodology is envisioned: the framework accelerates artificial intelligence/machine learning workloads and artificial intelligence/machine learning techniques enable the framework.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
人工智能和机器学习正在使基于实时数据的实时决策成为可能,用于交互式科学发现和自动驾驶和智能电网等使命关键应用。它们越来越多地由异构甚至可重新配置的加速器提供动力。加速器的可重新配置性和异构性,以及严格的性能要求和实时工作负载中的复杂依赖性,带来了令人生畏的运营挑战。这些问题如果得不到解决,将减缓科学发现,浪费大量的计算资源和能源。该项目将开发一个异构性和可重构性感知框架,以加速实时人工智能和机器学习,而不会损害其他工作负载。它将通过提高昂贵的计算系统的效率来造福社会,从而节省纳税人的钱并更好地利用现有投资。由该框架驱动的实时人工智能和机器学习可以更好地服务于社会,例如,加速科学发现并实现数据驱动的控制。该项目将为学术界和工业界的参与者带来创新的教育、推广和培训机会,为社会培养下一代研究人员和从业人员。今天,管理异构和可重构的系统,以满足不同的工作负载,同时提高资源利用率和性能保证是一项极具挑战性的任务。该项目将设计和实现一个自适应框架,该框架可以自动检测,分析和分析工作负载和加速器。根据这些信息,它自适应地重新配置它们,以使资源能力与工作负载需求相匹配。全局和局部优化将用于适应多种类型的工作负载以及每个工作负载中模型的配置、分区、放置、调度和执行。所开发的框架将提供可证明的性能,即使在未知的环境中的部分信息,这是迫切需要的,由于不断增加的系统复杂性和波动性的工作负载。本计画将以最佳化技术为基础,开发新的全球资源分配策略,以提供公平性、策略性及帕累托效率等效能保证。在整个项目中,设想了一种互惠的方法:框架加速人工智能/机器学习工作负载,人工智能/机器学习技术使框架成为可能。该奖项反映了NSF的法定使命,并通过使用基金会的智力价值和更广泛的影响审查标准进行评估,被认为值得支持。
项目成果
期刊论文数量(5)
专著数量(0)
科研奖励数量(0)
会议论文数量(0)
专利数量(0)
Deep Learning-Assisted Online Task Offloading for Latency Minimization in Heterogeneous Mobile Edge
- DOI:10.1109/tmc.2023.3285882
- 发表时间:2024-05
- 期刊:
- 影响因子:7.9
- 作者:Yu Liu;Yingling Mao;Z. Liu;Yuanyuan Yang
- 通讯作者:Yu Liu;Yingling Mao;Z. Liu;Yuanyuan Yang
Applied Online Algorithms with Heterogeneous Predictors
- DOI:
- 发表时间:2023
- 期刊:
- 影响因子:0
- 作者:Jessica Maghakian;Russell Lee;M. Hajiesmaili;Jian Li;R. Sitaraman;Zhenhu Liu
- 通讯作者:Jessica Maghakian;Russell Lee;M. Hajiesmaili;Jian Li;R. Sitaraman;Zhenhu Liu
Online Container Scheduling for Data-intensive Applications in Serverless Edge Computing
- DOI:10.1109/infocom53939.2023.10229034
- 发表时间:2023-05
- 期刊:
- 影响因子:0
- 作者:Xiaojun Shang;Yingling Mao;Yu Liu;Yaodong Huang;Zhen Liu;Yuanyuan Yang
- 通讯作者:Xiaojun Shang;Yingling Mao;Yu Liu;Yaodong Huang;Zhen Liu;Yuanyuan Yang
Energy-Aware Online Task Offloading and Resource Allocation for Mobile Edge Computing
- DOI:10.1109/icdcs57875.2023.00073
- 发表时间:2023-07
- 期刊:
- 影响因子:0
- 作者:Yu Liu;Yingling Mao;Xiaojun Shang;Z. Liu;Yuanyuan Yang
- 通讯作者:Yu Liu;Yingling Mao;Xiaojun Shang;Z. Liu;Yuanyuan Yang
Joint Task Offloading and Resource Allocation in Heterogeneous Edge Environments
- DOI:10.1109/infocom53939.2023.10229015
- 发表时间:2023-05
- 期刊:
- 影响因子:0
- 作者:Yu Liu;Yingling Mao;Z. Liu;Fan Ye;Yuanyuan Yang
- 通讯作者:Yu Liu;Yingling Mao;Z. Liu;Fan Ye;Yuanyuan Yang
{{
item.title }}
{{ item.translation_title }}
- DOI:
{{ item.doi }} - 发表时间:
{{ item.publish_year }} - 期刊:
- 影响因子:{{ item.factor }}
- 作者:
{{ item.authors }} - 通讯作者:
{{ item.author }}
数据更新时间:{{ journalArticles.updateTime }}
{{ item.title }}
- 作者:
{{ item.author }}
数据更新时间:{{ monograph.updateTime }}
{{ item.title }}
- 作者:
{{ item.author }}
数据更新时间:{{ sciAawards.updateTime }}
{{ item.title }}
- 作者:
{{ item.author }}
数据更新时间:{{ conferencePapers.updateTime }}
{{ item.title }}
- 作者:
{{ item.author }}
数据更新时间:{{ patent.updateTime }}
Zhenhua Liu其他文献
Copper-catalyzed C–N bond formation with imidazo[1,2-a]pyridines
铜催化咪唑并[1,2-a]吡啶形成 C–N 键
- DOI:
10.1039/c8ob01853g - 发表时间:
2018 - 期刊:
- 影响因子:0
- 作者:
Kai Sun;Shiqiang Mu;Zhenhua Liu;Ranran Feng;Yali Li;Kui Pang;Bing Zhang - 通讯作者:
Bing Zhang
From Darkness to Light: Pretargeted Radionuclide Imaging Driven by Tetrazine Bioorthogonal Chemistry
从黑暗到光明:四嗪生物正交化学驱动的预定位放射性核素成像
- DOI:
- 发表时间:
2018 - 期刊:
- 影响因子:3.4
- 作者:
Guohua Shen;Anren Kuang;Zhenhua Liu;Yige Bao;Haoxing Wu - 通讯作者:
Haoxing Wu
Experimental research on boiling heat transfer characteristics of compact staggered tube bundles in reduced pressures
- DOI:
10.3901/jme.2007.08.224 - 发表时间:
2007 - 期刊:
- 影响因子:4.2
- 作者:
Zhenhua Liu - 通讯作者:
Zhenhua Liu
Estimation of the homoplasmy degree for transplastomic tobacco using quantitative real-time PCR
使用定量实时 PCR 估计转质体烟草的同质性程度
- DOI:
10.1007/s00217-010-1265-z - 发表时间:
2010 - 期刊:
- 影响因子:3.3
- 作者:
Huifeng Shen;Bingjun Qian;Litao Yang;W. Liang;Weiwei Chen;Zhenhua Liu;Dabing Zhang - 通讯作者:
Dabing Zhang
An Asymptotic Formula in Number Theory
数论中的渐近公式
- DOI:
- 发表时间:
2016 - 期刊:
- 影响因子:0
- 作者:
Zhenhua Liu;Z. Dai - 通讯作者:
Z. Dai
Zhenhua Liu的其他文献
{{
item.title }}
{{ item.translation_title }}
- DOI:
{{ item.doi }} - 发表时间:
{{ item.publish_year }} - 期刊:
- 影响因子:{{ item.factor }}
- 作者:
{{ item.authors }} - 通讯作者:
{{ item.author }}
{{ truncateString('Zhenhua Liu', 18)}}的其他基金
Collaborative Research: CNS Core: Small: Optimizing Large-Scale Heterogeneous ML Platforms
合作研究:CNS Core:小型:优化大规模异构机器学习平台
- 批准号:
2146909 - 财政年份:2022
- 资助金额:
$ 53.31万 - 项目类别:
Standard Grant
Collaborative Research: CNS Core: Medium: Dynamic Data-driven Systems - Theory and Applications
合作研究:CNS 核心:媒介:动态数据驱动系统 - 理论与应用
- 批准号:
2106027 - 财政年份:2021
- 资助金额:
$ 53.31万 - 项目类别:
Standard Grant
NeTS: Small: Collaborative Research: Enabling Application-Level Performance Predictability in Public Clouds
NeTS:小型:协作研究:在公共云中实现应用程序级性能可预测性
- 批准号:
1617698 - 财政年份:2016
- 资助金额:
$ 53.31万 - 项目类别:
Standard Grant
CRII: NeTS: Enabling Demand Response from Cloud Data Centers -- from Sustainable IT to IT for Sustainability
CRII:NeTS:实现云数据中心的需求响应——从可持续 IT 到 IT 促进可持续发展
- 批准号:
1464388 - 财政年份:2015
- 资助金额:
$ 53.31万 - 项目类别:
Standard Grant
相似国自然基金
下一代无线通信系统自适应调制技术及跨层设计研究
- 批准号:60802033
- 批准年份:2008
- 资助金额:16.0 万元
- 项目类别:青年科学基金项目
由蝙蝠耳轮和鼻叶推导新型仿生自适应波束模型的研究
- 批准号:10774092
- 批准年份:2007
- 资助金额:39.0 万元
- 项目类别:面上项目
相似海外基金
CAREER: Adaptive Communications and Trajectory Design for UAV-assisted Wireless Networks: a Multi-Scale Decision Framework
职业:无人机辅助无线网络的自适应通信和轨迹设计:多尺度决策框架
- 批准号:
2129015 - 财政年份:2021
- 资助金额:
$ 53.31万 - 项目类别:
Continuing Grant
CAREER: Adaptive Communications and Trajectory Design for UAV-assisted Wireless Networks: a Multi-Scale Decision Framework
职业:无人机辅助无线网络的自适应通信和轨迹设计:多尺度决策框架
- 批准号:
2046034 - 财政年份:2021
- 资助金额:
$ 53.31万 - 项目类别:
Continuing Grant
CAREER: An Adaptive Stochastic Look-ahead Framework for Disaster Relief Logistics under Forecast Uncertainty
职业生涯:预测不确定性下救灾物流的自适应随机前瞻框架
- 批准号:
2045744 - 财政年份:2021
- 资助金额:
$ 53.31万 - 项目类别:
Standard Grant
CAREER: Applying a Criminological Framework to Understand Adaptive Adversarial Decision-Making Processes in Critical Infrastructure Cyberattacks
职业:应用犯罪学框架来理解关键基础设施网络攻击中的自适应对抗决策过程
- 批准号:
1453040 - 财政年份:2015
- 资助金额:
$ 53.31万 - 项目类别:
Continuing Grant
CAREER: A Human-Building Interaction Framework for Responsive and Adaptive Built Environments
职业:响应式和适应性建筑环境的人类建筑交互框架
- 批准号:
1351701 - 财政年份:2014
- 资助金额:
$ 53.31万 - 项目类别:
Standard Grant
CAREER: AIS - An Integrated Optimization and Prediction Framework for Machine Intelligence based on Adaptive Dynamic Programming
职业:AIS - 基于自适应动态规划的机器智能集成优化和预测框架
- 批准号:
1053717 - 财政年份:2011
- 资助金额:
$ 53.31万 - 项目类别:
Standard Grant
CAREER: From Nonstop-Monitoring to Nano-ISA: An Adaptive Multi-Dimensional Framework for Processor Reliability
职业生涯:从不间断监控到 Nano-ISA:处理器可靠性的自适应多维框架
- 批准号:
0954211 - 财政年份:2010
- 资助金额:
$ 53.31万 - 项目类别:
Continuing Grant
CAREER: Adaptive Architecture for Multicast Service Support in Large-Scale Mobile Ad Hoc Networks: Design and Evaluation Framework
职业:大规模移动自组织网络中多播服务支持的自适应架构:设计和评估框架
- 批准号:
0724658 - 财政年份:2006
- 资助金额:
$ 53.31万 - 项目类别:
Standard Grant
CAREER: A Receiver-Driven Framework for Scalable and Adaptive Peer-to-Peer Streaming
职业生涯:用于可扩展和自适应点对点流媒体的接收器驱动框架
- 批准号:
0448639 - 财政年份:2005
- 资助金额:
$ 53.31万 - 项目类别:
Continuing Grant
CAREER: Adaptive Architecture for Multicast Service Support in Large-Scale Mobile Ad Hoc Networks: Design and Evaluation Framework
职业:大规模移动自组织网络中多播服务支持的自适应架构:设计和评估框架
- 批准号:
0134650 - 财政年份:2002
- 资助金额:
$ 53.31万 - 项目类别:
Standard Grant