Skip to main content

德赢vwin官网客户端

Systems Research

URCS的计算机系统研究涵盖了一系德赢vwin官网客户端列主题,包括程序分析和编译器技术;并行,分布式和移动计算;基于群集的服务器技术;低功耗硬件和软件;处理器和内存架构;并发和同步;编程环境;和编程语言设计。

该部门的系统核心教师包括John CriswellChen DingSandhya dwarkadas.英国sreepathi pai.Michael Scott那和玉豪朱。In addition,Wendi Heinzelman那和迈克尔黄电气和计算机工程在CS中有联合约会,并积极参与部门研究。德赢vwin官网客户端

学院

John Criswell

John Cislwell的研究德赢vwin官网客户端兴趣专注于计算机安全和自动编译转换,可用于在商品软件上实施安全策略。他于2014年加入该部门在伊利诺伊州厄巴纳 - 香槟大学完成博士后。

Chen Ding

Chen Ding is the recipient of a DOE Early Career Principal Investigator award and an NSF CAREER award. He joined URCS in July, 2000, after receiving his PhD at Rice University. His research is in program analysis and optimization, including computational and mathematical models of locality and parallelism and automatic and suggestion-based techniques for locality optimization, memory management and program parallelization. He is widely known for his work on reuse distance to formally characterize locality and optimize data movement in the memory hierarchy of modern computer systems.

Sandhya dwarkadas.

Sandhya Dwarkadas是NSF博士后团契和NSF职业奖的收件人,并在软件分布式共享内存,并行架构和绩效评估方面做了工作。她联合领导了Cashmere交织分布式共享项目,并引导(Architecture, Run-time, and Compiler integration for High-performance computing) project.

sreepathi pai.

Sree is an experimental computer systems researcher interested in the performance of computer programs. To that end, he works in computer architecture, compilers and the implementation of programming languages.

Sree在由CPU和GPU组成的基于异构加速器的系统上。他最近的工作围绕了GPU上的高性能不规则/图形算法的优化编译器。Sree也是GPU上的图形算法的性能模型。

Michael Scott

Michael Scott是一家ACM研究员,IEEE研究员,以及大学的Goergen和Riker教学奖,Hajim学校终身成就奖,2006年的收件人Dijkstra Prize在分布式计算中。他的工作众所周知并发数据结构和同步算法。He co-led theCashmere交织分布式共享项目。他的教科书上编程语言设计和实现(编程语言语用学那Morgan Kaufmann, fourth edition, 2016) is a leading reference in the field, with adoptions at over 200 schools. His text on共享内存同步于2013年由摩根和克莱普浦发表。

玉豪朱

玉豪朱's research interests are computer architecture and software design to enable future mobile systems that are energy-efficient, intelligent, and offer desirable Quality-of-Experience. To that end, his recent work specifically focuses on domain-specific systems for visual computing (e.g., imaging, AR/VR, computer vision) and Web technologies.

Prior to join University of Rochester, he was a Visiting Researcher at ARM Research and a Research Fellow at Harvard University. He received his Ph.D. from UT Austin while being supported by the Google Ph.D. Fellowship.

Wendi Heinzelman

Wendi Heinzelman是NSF职业奖,INR年轻调查员奖,以及大学柯蒂斯教学奖的收件人。她于2001年1月开始到罗切斯特,在从麻省理工学院收到她的博士学位后。她的研究德赢vwin官网客户端涉及无线传感器网络和无线视频传输的算法和协议。她领导了这一点无线通信和网络group.

迈克尔黄

Michael Huang于2002年9月加入了该教师,在伊利诺伊大学的urbana-Champaign收到了他的博士后。他的兴趣位于计算机架构,处理器微体系结构,节能系统和处理器架构以及加工内存中。

自1974年成立以来,并行系统是该部门研究重点的中央部分。罗切斯特智能网关(钻机)项目的早期工作是CMU的强调和Mach项德赢vwin官网客户端目的直接前身,由Urum领导Rick Rashid(现在微软研究高级副总裁)。德赢vwin官网客户端这些反过来导致了商业上重要的操作系统作为Compaq Tru64和Apple Macos X.

在1980年代中期到末期,该部门的128 -点头e Butterfly Parallel Processor was the largest parallel computer anywhere in academia. It supported a wide variety of projects, including the Instant Replay debugging system, the Bridge parallel file system, the Elmwood and Psyche parallel operating systems, and the development of contention-free synchronization. Our work from this era is very heavily cited, and has influenced the research of many other groups.

Today our research spans a wide range of topics, including:

  • Program analysis and compiler technology
  • Parallel, distributed, and mobile computing
  • Cluster and cloud computing
  • 低功耗硬件和软件
  • 处理器和内存架构
  • Operating systems and security
  • 并发和同步

One of the strengths of the group is the way in which each faculty member's research interests tie into the others', allowing us to leverage infrastructure developed by the group as a whole. We therefore interact closely with the parallel user community, including researchers in astrophysics, data science, chemistry, biology, laser energetics, and, within our own department, computer vision, robotics, and machine learning. The multi-PI计算机系统和工程集团涵盖CS和ECE部门。

教授Dwarkadas的工作TreadMarksFastLink.projects, in collaboration with colleagues at Rice University and at the National Institutes of Health, was instrumental in discovering the gene believed responsible for Parkinson's disease. The work of former professor Kai Shen (now at Google) on cluster management was the technology behindAsk.com。来自斯科特教授组的几种算法出现在Java标准库中。

系统集团享有出色的计算资源。目前的研究基础设德赢vwin官网客户端施包括一个大型Linux集群,A 72线(36芯)英特尔·哈尔韦尔机,80螺纹(40芯)英特尔天窗机,160线(20芯)IBM Power8服务器,以及一个各种较小的X86机器。本集团还享有大学的各种机器Center for Integrated Research Computing,这激光能量实验室,这IBM Canada Centres for Advanced Studies和其他网站。

Project Pages

项目名 简要摘要
BOP: Behavior-oriented Parallelization aka Parallel Programming by Hints

Behavior-oriented parallelization (BOP) provides a suggestion interface for a user to mark possible parallelism and run-time support to guarante correctness and efficient execution whether the hints are correct or not. It enables program parallelization based on partial information and is useful for incrementally parallelizing a program or streamlining it for common uses.

Compiler Research, including Distance- and Footprint-Based Locality Analysis and Optimization

我们的编译器研究解决了正确德赢vwin官网客户端性和性能的双胞胎问题,重点是在使用数据确定复杂程序是否具有数据重用的固有模式,如果是的,则可以在多种程度上进行模拟,测量,修改(改进)。

COSYN:新兴多核处理器的通信和同步机制

该项目解决了使用组合的硬件软件方法的主流并行度的挑战。关键的想法是识别各种应用程序和编程模型的共同时间关键操作,可以通过新的架构机制加速或简化,然后将这些机制设计为普遍的时尚。候选机制包括警报更新通知机制,可编程数据隔离,自适应协作缓存和微粒访问控制。

共享内存并行程序的高性能同步

Synchronization serves to constrain the interleaving of actions performed by multiple threads of control (e.g., on a multicore processor), allowing only correct executions. Over the years, this ongoing project has developed some of the most efficient and widely used algorithms for locking, concurrent data structures, transactional memory, and persistence.

复杂系统性能建模与异常管理

该项目调查了用于多组件,数据密集型在线服务的配置文件驱动性能模型。它探讨了各种技巧,其中包括在Linux中识别出先前未知的I / O性能错误。

Reuse Distance and Footprint Based Locality Analysis and Optimization

重用距离和程序足迹是我们使用的两个基本度量,用于研究记忆系统性能和正确性的双胞胎问题,重点在使用数据时重新发生 - 确定复杂程序是否具有固有的数据重用模式,如果是的话,可以在多大程度上建模,测量和修改(改进)。

罗切斯特内存硬件错误研究项目德赢vwin官网客户端

此项目实时监控现场的计算机,并在发生时记录内存错误。它揭示了软(瞬态)误差是比以前报道的常数较少的数量级。它结合了软和硬(永久性)误差速率来预测整个系统的故障率和模式。

罗切斯特软件交易记忆(RSTM)系统

事务内存器(TM)允许程序员指定原因执行的操作,而不必担心如何实现原子性。RSTM下载到数千个地点,提供了多种套件的高效,相互兼容的TM运行时系统。

System Support for Fast Computational Accelerators

可编程GPU现在是跨越全频谱从手机到超级计算机的全频谱的节能计算。我们正在开发在相互不可信的应用程序中分享对这些和未来加速器的访问的技术,尽管具有高度可变的执行时间,但延长了托管CPU的能力的带宽要求以及可能绕过操作系统的通信信道。