• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文学位 >Many-task computing: Bridging the gap between high-throughput computing and high-performance computing.
【24h】

Many-task computing: Bridging the gap between high-throughput computing and high-performance computing.

机译：多任务计算：弥合高通量计算和高性能计算之间的差距。

获取原文
获取原文并翻译 | 示例
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Many-task computing aims to bridge the gap between two computing paradigms, high-throughput computing and high-performance computing. Many-task computing is reminiscent to high-throughput computing, but it differs in the emphasis of using many computing resources over short periods of time to accomplish many computational tasks, where the primary metrics are measured in seconds (e.g. tasks per second, I/O per second), as opposed to operations per month (e.g. jobs per month). Many-task computing denotes high-performance computations comprising of multiple distinct activities, coupled via file system operations. Tasks may be small or large, uniprocessor or multiprocessor, compute-intensive or data-intensive. The set of tasks may be static or dynamic, homogeneous or heterogeneous, loosely coupled or tightly coupled. The aggregate number of tasks, quantity of computing, and volumes of data may be extremely large. Many-task computing includes loosely coupled applications that are generally communication-intensive but not naturally expressed using message passing interface commonly found in high-performance computing, drawing attention to the many computations that are heterogeneous but not "happily" parallel. This dissertation explores fundamental issues in defining the many-task computing paradigm, as well as theoretical and practical issues in supporting both compute and data intensive many-task computing on large scale systems. We have defined an abstract model for data diffusion---an approach to supporting data-intensive many-task computing, have defined data-aware scheduling policies with heuristics to optimize real world performance, and developed a competitive online caching eviction policy. We also designed and implemented the necessary middleware---Falkon---to enable the support of many-task computing on clusters, grids and supercomputers. Micro-benchmarks have shown Falkon to achieve over 15K+ tasks/sec throughputs, scale to millions of queued tasks, to execute billions of tasks per day, and achieve hundreds of Gb/s I/O rates. Falkon has shown orders of magnitude improvements in performance and scalability across many diverse workloads (e.g. heterogeneous tasks from milliseconds to hours long, compute/data intensive, varying arrival rates) and applications (e.g. astronomy, medicine, chemistry, molecular dynamics, economic modeling, and data analytics) at scales of billions of tasks on hundreds of thousands of processors across Grids (e.g. TeraGrid) and supercomputers (e.g. IBM Blue Gene/P and Sun Constellation).
机译：多任务计算旨在弥合高吞吐量计算和高性能计算这两个计算范例之间的差距。多任务计算使人联想到高通量计算，但其重点在于在短时间内使用许多计算资源来完成许多计算任务，其中主要指标以秒为单位（例如，每秒的任务数，I /每秒O），而不是每月的操作（例如每月的工作）。多任务计算表示通过文件系统操作耦合的，包含多个不同活动的高性能计算。任务可以是小型或大型，单处理器或多处理器，计算密集型或数据密集型。任务集可以是静态的或动态的，同质的或异构的，松散耦合的或紧密耦合的。任务总数，计算量和数据量可能非常大。多任务计算包括松散耦合的应用程序，这些应用程序通常是通信密集型的，但不能使用高性能计算中常见的消息传递接口自然表达，从而引起人们对许多异构但非“愉快”并行计算的关注。本文探讨了定义多任务计算范式的基本问题，以及支持大规模系统上的计算和数据密集型多任务计算的理论和实践问题。我们已经定义了用于数据扩散的抽象模型-一种支持数据密集型多任务计算的方法，已经定义了具有启发式的数据感知调度策略以优化现实世界的性能，并开发了竞争性的在线缓存逐出策略。我们还设计并实现了必要的中间件-Falkon，以支持集群，网格和超级计算机上的多任务计算。微基准测试表明Falkon可实现超过15K +任务/秒的吞吐量，可扩展到数百万个排队任务，每天执行数十亿个任务以及达到数百Gb / s的I / O速率。 Falkon在许多不同的工作负载（例如，从毫秒到数小时的异构任务，计算/数据密集型，到达率变化）和应用程序（例如，天文学，医学，化学，分子动力学，经济模型，和数据分析）在网格（例如TeraGrid）和超级计算机（例如IBM Blue Gene / P和Sun Constellation）的数十万个处理器上完成数十亿个任务的规模。

著录项

  • 作者

    Raicu, Ioan.;

  • 作者单位

    The University of Chicago.;

  • 授予单位 The University of Chicago.;
  • 学科 Computer Science.
  • 学位 Ph.D.
  • 年度 2009
  • 页码 272 p.
  • 总页数 272
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类 自动化技术、计算机技术;
  • 关键词

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号