OPTIMIZING DATA PARTITIONING FOR DATA-PARALLEL COMPUTING
    1.
    发明申请
    OPTIMIZING DATA PARTITIONING FOR DATA-PARALLEL COMPUTING 有权
    优化用于数据并行计算的数据分区

    公开(公告)号:US20130152057A1

    公开(公告)日:2013-06-13

    申请号:US13325049

    申请日:2011-12-13

    IPC分类号: G06F9/44

    CPC分类号: G06F8/453

    摘要: A data partitioning plan is automatically generated that—given a data-parallel program and a large input dataset, and without having to first run the program on the input dataset—substantially optimizes performance of the distributed execution system that explicitly measures and infers various properties of both data and computation to perform cost estimation and optimization. Estimation may comprise inferring the cost of a candidate data partitioning plan, and optimization may comprise generating an optimal partitioning plan based on the estimated costs of computation and input/output.

    摘要翻译: 自动生成数据分区计划,给定数据并行程序和大型输入数据集,无需首先在输入数据集上运行程序,从而大大优化了分布式执行系统的性能,从而明确地测量和推断出 数据和计算都要进行成本估算和优化。 估计可以包括推断候选数据分割计划的成本,并且优化可以包括基于计算和输入/输出的估计成本来生成最优分割计划。

    Optimizing data partitioning for data-parallel computing
    2.
    发明授权
    Optimizing data partitioning for data-parallel computing 有权
    优化数据并行计算的数据分区

    公开(公告)号:US09235396B2

    公开(公告)日:2016-01-12

    申请号:US13325049

    申请日:2011-12-13

    IPC分类号: G06F9/44 G06F9/45

    CPC分类号: G06F8/453

    摘要: A data partitioning plan is automatically generated that—given a data-parallel program and a large input dataset, and without having to first run the program on the input dataset—substantially optimizes performance of the distributed execution system that explicitly measures and infers various properties of both data and computation to perform cost estimation and optimization. Estimation may comprise inferring the cost of a candidate data partitioning plan, and optimization may comprise generating an optimal partitioning plan based on the estimated costs of computation and input/output.

    摘要翻译: 自动生成数据分区计划,给定数据并行程序和大型输入数据集,无需首先在输入数据集上运行程序,从而大大优化了分布式执行系统的性能,从而明确地测量和推断出 数据和计算都要进行成本估算和优化。 估计可以包括推断候选数据分割计划的成本,并且优化可以包括基于计算和输入/输出的估计成本来生成最优分割计划。

    FINGERPRINTING EVENT LOGS FOR SYSTEM MANAGEMENT TROUBLESHOOTING
    3.
    发明申请
    FINGERPRINTING EVENT LOGS FOR SYSTEM MANAGEMENT TROUBLESHOOTING 有权
    指示事件日志用于系统管理故障排除

    公开(公告)号:US20100223499A1

    公开(公告)日:2010-09-02

    申请号:US12394451

    申请日:2009-02-27

    IPC分类号: G06F11/28 G06F11/07 G06F17/30

    摘要: A technique for automatically detecting and correcting configuration errors in a computing system. In a learning process, recurring event sequences, including e.g., registry access events, are identified from event logs, and corresponding rules are developed. In a detecting phase, the rules are applied to detected event sequences to identify violations and to recover from failures. Event sequences across multiple hosts can be analyzed. The recurring event sequences are identified efficiently by flattening a hierarchical sequence of the events such as is obtained from the Sequitur algorithm. A trie is generated from the recurring event sequences and edges of nodes of the trie are marked as rule edges or non-rule edges. A rule is formed from a set of nodes connected by rule edges. The rules can be updated as additional event sequences are analyzed. False positive suppression policies include a violation- consistency policy and an expected event disappearance policy.

    摘要翻译: 一种自动检测和纠正计算系统中配置错误的技术。 在学习过程中,从事件日志中识别循环事件序列,包括例如注册表访问事件,并且开发相应的规则。 在检测阶段,将规则应用于检测到的事件序列以识别违例行为并从故障中恢复。 可以分析多个主机的事件序列。 通过对诸如从Sequitur算法获得的事件的分层序列进行平坦化来有效地识别循环事件序列。 从循环事件序列生成特里(trie),并将特里斯的节点的边缘标记为规则边缘或非规则边缘。 规则是由一组通过规则边连接的节点形成的。 当分析附加事件序列时,可以更新规则。 虚假的积极抑制政策包括违规行为政策和预期的事件消失政策。

    Fingerprinting event logs for system management troubleshooting
    4.
    发明授权
    Fingerprinting event logs for system management troubleshooting 有权
    指纹事件日志用于系统管理故障排除

    公开(公告)号:US08069374B2

    公开(公告)日:2011-11-29

    申请号:US12394451

    申请日:2009-02-27

    IPC分类号: G06F11/00

    摘要: A technique for automatically detecting and correcting configuration errors in a computing system. In a learning process, recurring event sequences, including e.g., registry access events, are identified from event logs, and corresponding rules are developed. In a detecting phase, the rules are applied to detected event sequences to identify violations and to recover from failures. Event sequences across multiple hosts can be analyzed. The recurring event sequences are identified efficiently by flattening a hierarchical sequence of the events such as is obtained from the Sequitur algorithm. A trie is generated from the recurring event sequences and edges of nodes of the trie are marked as rule edges or non-rule edges. A rule is formed from a set of nodes connected by rule edges. The rules can be updated as additional event sequences are analyzed. False positive suppression policies include a violation-consistency policy and an expected event disappearance policy.

    摘要翻译: 一种自动检测和纠正计算系统中配置错误的技术。 在学习过程中,从事件日志中识别循环事件序列,包括例如注册表访问事件,并且开发相应的规则。 在检测阶段,将规则应用于检测到的事件序列以识别违例行为并从故障中恢复。 可以分析多个主机的事件序列。 通过对诸如从Sequitur算法获得的事件的分层序列进行平坦化来有效地识别循环事件序列。 从循环事件序列生成特里(trie),并将特里斯的节点的边缘标记为规则边缘或非规则边缘。 规则是由一组通过规则边连接的节点形成的。 当分析附加事件序列时,可以更新规则。 虚假的积极抑制政策包括违规一致性政策和预期的事件消失政策。

    Methods, systems, and media for protecting applications from races
    5.
    发明授权
    Methods, systems, and media for protecting applications from races 有权
    用于保护应用程序免于种族的方法,系统和媒体

    公开(公告)号:US09038048B2

    公开(公告)日:2015-05-19

    申请号:US13189263

    申请日:2011-07-22

    IPC分类号: G06F9/44 G06F9/52

    CPC分类号: G06F9/52 G06F8/65

    摘要: Methods, systems, and media for protecting applications from races are provided. In some embodiments, methods for protecting applications from races are provided, the methods comprising: adding to at least one cycle of the application an update check to determine when an update to the application is to be made; adding an update engine to the application, wherein the update engine is configured to: receive an update plan that is based on an execution filter that specifies how operations of the application are to be synchronized; and cause synchronization operations to be added to the application based on the update plan that prevent the race from occurring.

    摘要翻译: 提供了用于保护应用程序免于种族的方法,系统和媒体。 在一些实施例中,提供了用于保护应用程序免于种族的方法,所述方法包括:向应用程序的至少一个周期添加更新检查,以确定何时进行对应用的更新; 向所述应用添加更新引擎,其中所述更新引擎被配置为:接收基于执行过滤器的更新计划,所述执行过滤器指定所述应用的操作如何被同步; 并且基于防止发生竞争的更新计划使得将同步操作添加到应用程序。

    Methods, Systems, and Media for Protecting Applications from Races
    6.
    发明申请
    Methods, Systems, and Media for Protecting Applications from Races 有权
    用于保护应用程序的方法,系统和媒体

    公开(公告)号:US20120096449A1

    公开(公告)日:2012-04-19

    申请号:US13189263

    申请日:2011-07-22

    IPC分类号: G06F9/44

    CPC分类号: G06F9/52 G06F8/65

    摘要: Methods, systems, and media for protecting applications from races are provided. In some embodiments, methods for protecting applications from races are provided, the methods comprising: adding to at least one cycle of the application an update check to determine when an update to the application is to be made; adding an update engine to the application, wherein the update engine is configured to: receive an update plan that is based on an execution filter that specifies how operations of the application are to be synchronized; and cause synchronization operations to be added to the application based on the update plan that prevent the race from occurring.

    摘要翻译: 提供了用于保护应用程序免于种族的方法,系统和媒体。 在一些实施例中,提供了用于保护应用程序免于种族的方法,所述方法包括:向应用程序的至少一个周期添加更新检查,以确定何时进行对应用的更新; 向所述应用添加更新引擎,其中所述更新引擎被配置为:接收基于执行过滤器的更新计划,所述执行过滤器指定所述应用的操作如何被同步; 并且基于防止发生竞争的更新计划使得将同步操作添加到应用程序。

    Methods, Systems, and Media for Providing Determinism in Multithreaded Programs
    7.
    发明申请
    Methods, Systems, and Media for Providing Determinism in Multithreaded Programs 有权
    在多线程程序中提供确定性的方法,系统和媒体

    公开(公告)号:US20120102470A1

    公开(公告)日:2012-04-26

    申请号:US13190316

    申请日:2011-07-25

    IPC分类号: G06F9/44

    CPC分类号: G06F11/3632 G06F8/458

    摘要: Methods, systems, and media for providing determinism in multithreaded programs are provided. In some embodiments, methods for providing determinism in multithreaded programs are provided, the methods comprising: instrumenting a program with: a proxy that: receives inputs to the program; and determines whether a schedule corresponding to a set of input constraints satisfied by the inputs has been memoized; and a replayer that: if it is determined that a schedule corresponding to a set of input constraints satisfied by the inputs has been memoized, processes the inputs by the program according to the schedule.

    摘要翻译: 提供了在多线程程序中提供确定性的方法,系统和媒体。 在一些实施例中,提供了用于在多线程程序中提供确定性的方法,所述方法包括:使用以下代理来对程序进行测试:代理,其接收对所述程序的输入; 并且确定与输入满足的一组输入约束对应的调度是否被记忆; 以及重播者:如果确定与输入满足的一组输入约束对应的日程表已被记录,则根据该日程表处理该程序的输入。

    Methods, systems, and media for providing determinism in multithreaded programs
    8.
    发明授权
    Methods, systems, and media for providing determinism in multithreaded programs 有权
    用于在多线程程序中提供确定性的方法,系统和媒体

    公开(公告)号:US09454460B2

    公开(公告)日:2016-09-27

    申请号:US13190316

    申请日:2011-07-25

    IPC分类号: G06F9/46 G06F11/36 G06F9/45

    CPC分类号: G06F11/3632 G06F8/458

    摘要: Methods, systems, and media for providing determinism in multithreaded programs are provided. In some embodiments, methods for providing determinism in multithreaded programs are provided, the methods comprising: instrumenting a program with: a proxy that: receives inputs to the program; and determines whether a schedule corresponding to a set of input constraints satisfied by the inputs has been memoized; and a replayer that: if it is determined that a schedule corresponding to a set of input constraints satisfied by the inputs has been memoized, processes the inputs by the program according to the schedule.

    摘要翻译: 提供了在多线程程序中提供确定性的方法,系统和媒体。 在一些实施例中,提供了用于在多线程程序中提供确定性的方法,所述方法包括:使用以下代理来对程序进行测试:代理:接收对所述程序的输入; 并且确定与输入满足的一组输入约束对应的日程表是否被记忆; 以及重播者:如果确定与输入满足的一组输入约束对应的日程表已被记录,则根据该日程表处理该程序的输入。

    Methods, Systems, and Media for Protecting Applications from Races
    9.
    发明申请
    Methods, Systems, and Media for Protecting Applications from Races 审中-公开
    用于保护应用程序的方法,系统和媒体

    公开(公告)号:US20150254112A1

    公开(公告)日:2015-09-10

    申请号:US14681872

    申请日:2015-04-08

    IPC分类号: G06F9/52 G06F9/445

    CPC分类号: G06F9/52 G06F8/65

    摘要: Methods, systems, and media for protecting applications from races are provided. In some embodiments, methods for protecting applications from races are provided, the methods comprising: adding to at least one cycle of the application an update check to determine when an update to the application is to be made; adding an update engine to the application, wherein the update engine is configured to: receive an update plan that is based on an execution filter that specifies how operations of the application are to be synchronized; and cause synchronization operations to be added to the application based on the update plan that prevent the race from occurring.

    摘要翻译: 提供了用于保护应用程序免于种族的方法,系统和媒体。 在一些实施例中,提供了用于保护应用程序免于种族的方法,所述方法包括:向应用程序的至少一个周期添加更新检查,以确定何时进行对应用的更新; 向所述应用添加更新引擎,其中所述更新引擎被配置为:接收基于执行过滤器的更新计划,所述执行过滤器指定所述应用的操作如何被同步; 并且基于防止发生竞争的更新计划使得将同步操作添加到应用程序。

    Distributed system checker
    10.
    发明授权
    Distributed system checker 有权
    分布式系统检查器

    公开(公告)号:US07984332B2

    公开(公告)日:2011-07-19

    申请号:US12272779

    申请日:2008-11-17

    IPC分类号: G06F11/00

    CPC分类号: G06F11/362

    摘要: A distributed system checker may check a distributed system against events to detect bugs in the distributed system. The events may include machines crashes, network partitions, and packet losses, for example. The distributed system checker may check a distributed system that can have multiple threads and multiple processes running on multiple nodes. To obtain control over a distributed system, a distributed system checker may insert an interposition layer between a process and the operating system on each node.

    摘要翻译: 分布式系统检查器可以根据事件检查分布式系统以检测分布式系统中的错误。 例如,事件可能包括机器崩溃,网络分区和数据包丢失。 分布式系统检查器可以检查可以在多个节点上运行多个线程和多个进程的分布式系统。 为了获得对分布式系统的控制,分布式系统检查器可以在进程和每个节点上的操作系统之间插入插入层。