SHF: Medium: Collaborative Research: Machine Learning Enabled Network-on-Chip Architectures for Optimized Energy, Performance and Reliability
SHF: Medium: Collaborative Research: Machine Learning Enabled Network-on-Chip Architectures for Optimized Energy, Performance and Reliability
批准号:
1703013
负责人:
Avinash Karanth
金额:
$50.0万
依托单位:
依托单位国家:
美国
项目类别:
Continuing Grant
财政年份:
2017
资助国家:
美国
项目状态:
已结题
起止时间:
2017-06-01 至 2023-09-30
中文摘要
片上网络(NoC)体系结构已经成为多核和芯片多处理器(CMPS)的主流片上通信结构。然而,随着NOC架构的扩展,它们面临着严峻的挑战。当今解决优化的NoC架构设计的一个关键挑战是NoC设计者可以使用的过多的性能增强、节能和容错技术,以及必须导航的巨大设计空间,以同时降低功耗、提高可靠性、提高性能和维护服务质量。这项研究提出了一种新的跨层、横切的方法,横跨电路、体系结构、机器学习算法和应用程序,旨在设计节能、可靠和可扩展的片上网络。这项研究将产生(1)新颖的跨层设计技术,这些技术采取全面的方法来同时降低功耗,同时仍然实现NoC的可靠性和性能目标,(2)基本了解如何使用硬件兼容ML进行NoC设计优化,(3)用于监控和收集网络执行期间的关键数据和关键设计参数以优化NoC设计的软件和硬件技术,以及(4)将改进体系结构社区用于评估可扩展NoC设计方法的建模和仿真工具。拟议的研究在硬件架构师和机器学习科学家之间架起了一座非常重要的鸿沟,前者在电路和体系结构级别设计电源管理和容错技术,后者开发预测和优化技术。由于其交叉性质,拟议的研究有可能显著改变下一代CMPS和片上系统(SoC)的设计,在这些领域,必须做出影响功率、性能和可靠性的复杂决策。这项研究还将通过将发现与教学和培训相结合,在教育中发挥重要作用。作为拟议项目的一部分,私人投资机构致力于并将继续扩大外联活动,作出必要努力吸引和培训这一领域的少数族裔学生。
英文摘要
Network-on-Chip (NoC) architectures have emerged as the prevailing on-chip communication fabric for multicores and Chip Multiprocessors (CMPs). However, as NoC architectures are scaled, they face serious challenges. A key challenge in addressing optimized NoC architecture design today is the plethora of performance enhancing, energy efficient and fault tolerant techniques available to NoC designers and the large design space that must be navigated to simultaneously reduce power, improve reliability, increase performance and maintain QoS. This research proposes a new cross-layer, cross-cutting methodology spanning circuits, architectures, machine learning algorithms, and applications, aimed at designing energy-efficient, reliable and scalable NoCs. This research will result in (1) novel cross-layer design techniques that take a holistic approach of simultaneously reducing power consumption, while still achieving reliability and performance goals for NoCs, (2) a fundamental understanding of the use of hardware-amenable ML for NoC design optimization, (3) software and hardware techniques for monitoring and collecting critical data and key design parameters during network execution to optimize NoC design, and (4) modeling and simulation tools that will improve the architecture community's design methodologies for evaluating scalable NoCs. The proposed research bridges a very important gap between hardware architects who design power management and fault tolerant techniques at the circuit and architecture level and machine learning scientists who develop predictive and optimization techniques. Due to its cross-cutting nature, the proposed research has the potential to significantly transform the design of next-generation CMPs and System-on-Chips (SoCs) where complex decisions have to be made that affect the power, performance and reliability. The research will also play a major role in education by integrating discovery with teaching and training. The PIs are committed and will continue to expand on outreach activities as part of the proposed project by making the necessary efforts to attract and train minority students in this field.
期刊论文(17)
专著(0)
科研奖励(0)
会议论文
登录
查看更多内容
Parallel Dot Products Using Silicon Photonics
使用硅光子学的并行点积
DOI:
10.1109/ipc48725.2021.9592936
发表时间:
2021
期刊:
2021 IEEE Photonics Conference (IPC
影响因子:
--
作者:
[Wolff, Andy, Shiflett, Kyle, Karanth, Avinash]
通讯作者:
Karanth, Avinash
DOI:
10.1145/3453688.3461515
发表时间:
2021-06
期刊:
Proceedings of the 2021 Great Lakes Symposium on VLSI
影响因子:
--
作者:
[Kyle Shiflett;Avinash Karanth;A. Louri;Razvan C. Bunescu]
通讯作者:
Kyle Shiflett;Avinash Karanth;A. Louri;Razvan C. Bunescu
DOI:
10.1109/tpds.2023.3327535
发表时间:
2024-01
期刊:
IEEE Transactions on Parallel and Distributed Systems
影响因子:
5.3
作者:
[Yuan Li;A. Louri;Avinash Karanth]
通讯作者:
Yuan Li;A. Louri;Avinash Karanth
DOI:
10.1109/tc.2018.2875476
发表时间:
2019-03-01
期刊:
IEEE TRANSACTIONS ON COMPUTERS
影响因子:
3.7
作者:
[Fettes, Quintin, Clark, Mark, Louri, Ahmed]
通讯作者:
Louri, Ahmed
Hardware-Level Thread Migration to Reduce On-Chip Data Movement Via Reinforcement Learning
通过强化学习进行硬件级线程迁移以减少片上数据移动
DOI:
10.1109/tcad.2020.3012650
发表时间:
2020
期刊:
IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
影响因子:
2.9
作者:
[Fettes, Quintin, Karanth, Avinash, Bunescu, Razvan, Louri, Ahmed, Shiflett, Kyle]
通讯作者:
Shiflett, Kyle
共 17 条
Collaborative Research: DESC: Type II: Multi-Function Cross-Layer Electro-Optic Fabrics for Reliable and Sustainable Computing Systems
-
批准号:2324645
-
项目类别:Standard Grant
-
资助金额:$100.0万
-
财政年份:2023
-
负责人:Avinash Karanth
-
依托单位:
Collaborative Research: SHF: Medium: EPIC: Exploiting Photonic Interconnects for Resilient Data Communication and Acceleration in Energy-Efficient Chiplet-based Architectures
-
批准号:2311544
-
项目类别:Continuing Grant
-
资助金额:$60.0万
-
财政年份:2023
-
负责人:Avinash Karanth
-
依托单位:
SaTC: CORE: Small: Language Abstractions for Reconfigurable Hardware Monitors on Manycore Architectures
-
批准号:1936794
-
项目类别:Standard Grant
-
资助金额:$49.94万
-
财政年份:2020
-
负责人:Avinash Karanth
-
依托单位:
SHF: Medium: Collaborative Research: Photonic Neural Network Accelerator for Energy-efficient Heterogeneous Multicore Architectures
-
批准号:1901192
-
项目类别:Continuing Grant
-
资助金额:$50.0万
-
财政年份:2019
-
负责人:Avinash Karanth
-
依托单位:
SHF: Medium: Collaborative Research: Scaling On-chip Networks to 1000-core Systems using Heterogeneous Emerging Interconnect Technologies
-
批准号:1513606
-
项目类别:Continuing Grant
-
资助金额:$48.0万
-
财政年份:2015
-
负责人:Avinash Karanth
-
依托单位:
SHF: Small: Collaborative Research: A Holistic Design Methodology for Fault-Tolerant and Robust Network-on-Chips (NoCs) Architectures
-
批准号:1420718
-
项目类别:Standard Grant
-
资助金额:$20.0万
-
财政年份:2014
-
负责人:Avinash Karanth
-
依托单位:
SHF: Small: Collaborative Research: Power-Efficient and Reliable 3D Stacked Reconfigurable Photonic Network-on-Chips for Scalable Multicore Architectures
-
批准号:1318981
-
项目类别:Standard Grant
-
资助金额:$14.5万
-
财政年份:2013
-
负责人:Avinash Karanth
-
依托单位:
Collaborative Research:EAGER:Exploiting Heterogeneity in Emerging Interconnect Technologies for Building Highly Scalable and Power-Efficient Network-on-Chips for Many-core Systems
-
批准号:1342657
-
项目类别:Standard Grant
-
资助金额:$5.0万
-
财政年份:2013
-
负责人:Avinash Karanth
-
依托单位:
Power-Efficient Reconfigurable Wireless Network-on-Chip (NoC) Interconnects for Future Many-core Architectures
-
批准号:1129010
-
项目类别:Standard Grant
-
资助金额:$38.0万
-
财政年份:2011
-
负责人:Avinash Karanth
-
依托单位:
CAREER: Design of Reconfigurable Power and Area-Efficient Nanophotonic Architectures for Future Multi-cores
-
批准号:1054339
-
项目类别:Continuing Grant
-
资助金额:$40.76万
-
财政年份:2011
-
负责人:Avinash Karanth
-
依托单位:
SHF: Small: Collaborative Research: Design of Power and Area Efficient, Fault-tolerant Network-on-Chip Circuits and Architectures
-
批准号:0915418
-
项目类别:Standard Grant
-
资助金额:$13.44万
-
财政年份:2009
-
负责人:Avinash Karanth
-
依托单位:
海外基金