课题基金 / 基金详情

Thermal-Aware Multiprocessor System-On-chip Synthesis

Thermal-Aware Multiprocessor System-On-chip Synthesis
热感知多处理器片上系统综合
批准号:
0964763
负责人:
Robert Dick
金额:
$21.0万
依托单位国家:
美国
项目类别:
Standard Grant
财政年份:
2009
资助国家:
美国
项目状态:
已结题
起止时间:
2009-05-15 至 2012-04-30

项目摘要

项目成果

Robert Dick的其他基金

相似基金

相关文献

中文摘要
翻译
摘要:嵌入式系统是应用于其他设备的专用计算机,如消费电子产品、汽车、医疗设备或用于监测建筑物安全和野生动物健康的传感器网络。许多嵌入式系统现在几乎完全在一个多处理器片上系统(MPSoC)上实现,这是一个包含复杂微处理器网络的单个集成电路芯片。在高性能应用中,限制计算速度的关键因素是温度。设计mpsoc是一项艰巨的任务,因为每个应用程序都有不同的要求。在这个项目中,pi计划开发准确预测MPSoC设计决策对温度、功能、价格和功耗影响的方法。我们还将开发自动设计适合特定应用的mpsoc的方法。计划中的工作将侧重于可以在许多不同的MPSoC应用中广泛重用的想法。据估计,到2007年,片上系统市场预计将达到1700亿美元。通过与我们的工业分支机构在设计项目上的合作,并通过向未来将加入工业界的学生传授新的设计技术,该项目有望鼓励工业界采用本项目中开发的最有前途的新设计和自动化设计技术。
英文摘要
Proposal ID: 0702761PI name: Robert DickTitle: Thermal-Aware Multiprocessor System-On-chip SynthesisInstitution: Northwestern UniversityABSTRACT:Embedded systems are application-specific computers within other devices such as consumer electronics, automobiles, medical devices, or networks of sensors used to monitor the safety of buildings and the health of wildlife. Many embedded systems are now implemented almost entirely on a multiprocessor system-on-chip (MPSoC), a single integrated circuit chip containing a complex network of microprocessors. In high-performance applications, the critical limiter on computation speed is temperature. Designing MPSoCs is a difficult task because the each application imposes different requirements. In this project, the PIs plan to develop methods of accurately predicting the impacts of MPSoC design decisions on temperature, functionality, price, and power consumption. We will also develop methods of automatically designing MPSoCs that are well suited to particular applications. The planned work will focus on ideas that can be widely re-used in many different MPSoC applications. An estimate is that system-on-chip market is projected to reach $170 billion by 2007. The project will hopefully work to encourage industry to adopt the most promising new design and automated design techniques developed in this project by collaborating with our industrial affiliates on design projects and by teaching new design techniques to students who will later join industry.
期刊论文(0)
专著(0)
科研奖励(0)
会议论文
Collaborative Research: CNS Core: Medium: The Privacy Backplane - A Full Stack Approach to Individualized Privacy Controls Throughout the Internet-of-Things
CNS Core: Small: Importance-Aware Compressive Inference for Efficient Embedded Vision
I-Corps: Blocking-, Censorship-, Surveillance-, and Disaster-Resistant Communication for Normal Smartphone Users
CyberSEES: Type 2: Collaborative Research: Connecting Next-generation Air Pollution Exposure Measurements to Environmentally Sustainable Communities
海外基金