公开(公告)号：US20080168249A1

公开(公告)日：2008-07-10

申请号：US11620818

申请日：2007-01-08

申请人： Martin Hirzel ,  Chengliang Zhang

发明人： Martin Hirzel ,  Chengliang Zhang

IPC分类号： G06F12/02

CPC分类号： G06F12/0253

摘要： Disclosed are a method and system for finding an optimal data layout. The approach of the present invention is to try one of several data layouts in the memory, measure the impact of said one data layout on a performance of a program, and decide which of said several data layouts to try next. The trying and measuring steps are repeated, and one of said several data layouts is selected as best or optimal based on the measurings. The preferred embodiment of the invention provides layout auditing, a framework for picking the best data layout online without requiring any user input. Layout auditing optimizes data layouts with a try-measure-decide feedback loop: use a data reorganizer to try one of several data layouts, use a profiler to measure the impact of the data layout on performance, and use a controller to decide which data layout to try next.

摘要翻译： 公开了一种用于找到最佳数据布局的方法和系统。 本发明的方法是尝试存储器中的几个数据布局中的一个，测量所述一个数据布局对程序的性能的影响，并且决定接下来要尝试的几个数据布局中的哪一个。 重复尝试和测量步骤，并且基于测量将所述多个数据布局中的一个选择为最佳或最佳。 本发明的优选实施例提供布局审核，用于在线选择最佳数据布局的框架，而不需要任何用户输入。 布局审核通过测试决定反馈循环优化数据布局：使用数据重组器尝试几个数据布局之一，使用分析器来衡量数据布局对性能的影响，并使用控制器来决定哪些数据布局 尝试下一个

公开(公告)号：US20070083856A1

公开(公告)日：2007-04-12

申请号：US11539111

申请日：2006-10-05

申请人： Trishul Chilimbi ,  Martin Hirzel

发明人： Trishul Chilimbi ,  Martin Hirzel

IPC分类号： G06F9/44

CPC分类号： G06F11/3612

摘要： A temporal profiling framework useful for dynamic optimization with hot data stream prefetching provides profiling of longer bursts and lower overhead. For profiling longer bursts, the framework employs a profiling phase counter, as well as a checking phase counter, to control transitions to and from instrumented code for sampling bursts of a program execution trace. The temporal profiling framework further intelligently eliminates some checks at procedure entries and loop back-edges, while still avoiding unbounded execution without executing checks for transition to and from instrumented code. Fast hot data stream detection analyzes a grammar of a profiled data reference sequence, calculating a heat metric for recurring subsequences based on length and number of unique occurrences outside of other hot data streams in the sequence with sufficiently low-overhead to permit use in a dynamic optimization framework.

摘要翻译： 用于使用热数据流预取的动态优化的时间分析框架提供了更长的突发和较低开销的分析。 为了分析更长的脉冲串，框架采用分析相位计数器以及检查相位计数器，以控制到用于对程序执行跟踪的脉冲串进行采样的代码的转换。 时间分析框架进一步智能地消除了过程条目和循环后沿的一些检查，同时仍然避免了无限制的执行，而无需执行向和从仪器化代码转换的检查。 快速热数据流检测分析分析数据参考序列的语法，基于序列中其他热数据流之外的独特事件的长度和数量计算循环子序列的热度量，具有足够低的开销以允许在动态 优化框架。

公开(公告)号：US07140008B2

公开(公告)日：2006-11-21

申请号：US10305056

申请日：2002-11-25

申请人： Trishul A. Chilimbi ,  Martin Hirzel

发明人： Trishul A. Chilimbi ,  Martin Hirzel

IPC分类号： G06F9/45

CPC分类号： G06F11/3612

摘要： A temporal profiling framework useful for dynamic optimization with hot data stream prefetching provides profiling of longer bursts and lower overhead. For profiling longer bursts, the framework employs a profiling phase counter, as well as a checking phase counter, to control transitions to and from instrumented code for sampling bursts of a program execution trace. The temporal profiling framework further intelligently eliminates some checks at procedure entries and loop back-edges, while still avoiding unbounded execution without executing checks for transition to and from instrumented code. Fast hot data stream detection analyzes a grammar of a profiled data reference sequence, calculating a heat metric for recurring subsequences based on length and number of unique occurrences outside of other hot data streams in the sequence with sufficiently low-overhead to permit use in a dynamic optimization framework.

摘要翻译： 用于使用热数据流预取的动态优化的时间分析框架提供了更长的突发和较低开销的分析。 为了分析更长的脉冲串，框架采用分析相位计数器以及检查相位计数器，以控制到用于对程序执行跟踪的脉冲串进行采样的代码的转换。 时间分析框架进一步智能地消除了过程条目和循环后沿的一些检查，同时仍然避免了无限制的执行，而无需执行向和从仪器化代码转换的检查。 快速热数据流检测分析分析数据参考序列的语法，基于序列中其他热数据流之外的独特事件的长度和数量计算循环子序列的热度量，具有足够低的开销以允许在动态 优化框架。

公开(公告)号：US07930325B2

公开(公告)日：2011-04-19

申请号：US12129067

申请日：2008-05-29

申请人： David K. Siegwart ,  Martin Hirzel

发明人： David K. Siegwart ,  Martin Hirzel

IPC分类号： G06F17/30 ,  G06F17/00

CPC分类号： G06F12/0253

摘要： A garbage collection algorithm that achieves hierarchical copy order with parallel garbage collection threads. More specifically, the present invention provides a garbage collection method and system for copying objects from a from-space to a to-space. The method comprises the steps of (a) having multiple threads that simultaneously perform work for garbage collection (GC), (b) examining the placement of objects on blocks, and (c) changing the placement of objects on blocks based on step (b). Preferably, the method includes the additional step of calculating a placement of object(s) based on step (b), and using the result of the calculation for step (c). For example, the calculation may be used to increase the frequency of intra-block pointers and/or to increase the frequency of siblings on the same block.

摘要翻译： 一种垃圾收集算法，可以实现具有并行垃圾收集线程的分层复制顺序。 更具体地说，本发明提供了一种用于从空间到空间复制对象的垃圾收集方法和系统。 该方法包括以下步骤：（a）具有同时执行垃圾回收（GC）的工作的多个线程，（b）检查对象在块上的放置，以及（c）基于步骤（b）改变对象在块上的放置 ）。 优选地，该方法包括基于步骤（b）计算物体的放置并使用步骤（c）的计算结果的附加步骤。 例如，该计算可以用于增加块内指针的频率和/或增加同一块上的兄弟姐妹的频率。

公开(公告)号：US20080235307A1

公开(公告)日：2008-09-25

申请号：US12129049

申请日：2008-05-29

申请人： David K. Siegwart ,  Martin Hirzel

发明人： David K. Siegwart ,  Martin Hirzel

IPC分类号： G06F12/00 ,  G06F17/30

CPC分类号： G06F12/0253

摘要： Disclosed is a garbage collection algorithm that achieves hierarchical copy order with parallel garbage collection threads. More specifically, the present invention provides a garbage collection method and system for copying objects from a from-space to a to-space. The method comprises the steps of (a) having multiple threads that simultaneously perform work for garbage collection (GC), (b) examining the placement of objects on blocks, and (c) changing the placement of objects on blocks based on step (b). Preferably, the method includes the additional step of calculating a placement of object(s) based on step (b), and using the result of the calculation for step (c). For example, the calculation may be used to increase the frequency of intra-block pointers and/or to increase the frequency of siblings on the same block.

摘要翻译： 公开了一种利用并行垃圾收集线程实现分层复制顺序的垃圾收集算法。 更具体地说，本发明提供了一种用于从空间到空间复制对象的垃圾收集方法和系统。 该方法包括以下步骤：（a）具有同时执行垃圾回收（GC）的工作的多个线程，（b）检查对象在块上的放置，以及（c）基于步骤（b）改变对象在块上的放置 ）。 优选地，该方法包括基于步骤（b）计算物体的放置并使用步骤（c）的计算结果的附加步骤。 例如，该计算可以用于增加块内指针的频率和/或增加同一块上的兄弟姐妹的频率。

公开(公告)号：US08046752B2

公开(公告)日：2011-10-25

申请号：US11280476

申请日：2005-11-15

申请人： Trishul Chilimbi ,  Martin Hirzel

发明人： Trishul Chilimbi ,  Martin Hirzel

IPC分类号： G06F9/45 ,  G06F9/26 ,  G06F9/34 ,  G06F9/30 ,  G06F7/40

CPC分类号： G06F9/45525 ,  G06F8/4442

摘要： A method and system for creating and injecting code into a running program that identifies a hot data stream, and prefetching data elements in the stream so they are available when needed by the processor. The injected code identifies the first few elements in a hot data stream (i.e. the prefix), and prefetches the balance of the elements in the stream (i.e., the suffix). Since the hot data stream identification code and prefetch code is injected at run time, pointer related time-dependencies inherent in earlier prefetch systems are eliminated. A global deterministic finite state machine (DFSM) is used to help create conceptual logic used to generate the code injected into the program for prefix detection.

摘要翻译： 一种方法和系统，用于创建和注入代码以识别热数据流的运行程序，以及预取数据流中的数据元素，使其在处理器需要时可用。 注入的代码识别热数据流中的前几个元素（即前缀），并且预取流中元素的平衡（即，后缀）。 由于在运行时注入了热数据流识别码和预取代码，消除了早期预取系统固有的指针相关时间依赖性。 全局确定性有限状态机（DFSM）用于帮助创建用于生成注入到程序中用于前缀检测的代码的概念逻辑。

公开(公告)号：US20080250089A1

公开(公告)日：2008-10-09

申请号：US12128684

申请日：2008-05-29

申请人： David K. Siegwart ,  Martin Hirzel

发明人： David K. Siegwart ,  Martin Hirzel

IPC分类号： G06F17/30

CPC分类号： G06F12/0253

摘要： Disclosed is a garbage collection algorithm that achieves hierarchical copy order with parallel garbage collection threads. More specifically, the present invention provides a garbage collection method and system for copying objects from a from-space to a to-space. The method comprises the steps of (a) having multiple threads that simultaneously perform work for garbage collection (GC), (b) examining the placement of objects on blocks, and (c) changing the placement of objects on blocks based on step (b). Preferably, the method includes the additional step of calculating a placement of object(s) based on step (b), and using the result of the calculation for step (c). For example, the calculation may be used to increase the frequency of intra-block pointers and/or to increase the frequency of siblings on the same block.

摘要翻译： 公开了一种利用并行垃圾收集线程实现分层复制顺序的垃圾收集算法。 更具体地说，本发明提供了一种用于从空间到空间复制对象的垃圾收集方法和系统。 该方法包括以下步骤：（a）具有同时执行垃圾回收（GC）的工作的多个线程，（b）检查对象在块上的放置，以及（c）基于步骤（b）改变对象在块上的放置 ）。 优选地，该方法包括基于步骤（b）计算物体的放置并使用步骤（c）的计算结果的附加步骤。 例如，该计算可以用于增加块内指针的频率和/或增加同一块上的兄弟姐妹的频率。

公开(公告)号：US20060070040A1

公开(公告)日：2006-03-30

申请号：US11280476

申请日：2005-11-15

申请人： Trishul Chilimbi ,  Martin Hirzel

发明人： Trishul Chilimbi ,  Martin Hirzel

IPC分类号： G06F9/44

CPC分类号： G06F9/45525 ,  G06F8/4442

摘要： A method and system for creating and injecting code into a running program that identifies a hot data stream, and prefetching data elements in the stream so they are available when needed by the processor. The injected code identifies the first few elements in a hot data stream (i.e. the prefix), and prefetches the balance of the elements in the stream (i.e., the suffix). Since the hot data stream identification code and prefetch code is injected at run time, pointer related time-dependencies inherent in earlier prefetch systems are eliminated. A global deterministic finite state machine (DFSM) is used to help create conceptual logic used to generate the code injected into the program for prefix detection.

摘要翻译： 一种方法和系统，用于创建和注入代码以识别热数据流的运行程序，以及预取数据流中的数据元素，使其在处理器需要时可用。 注入的代码识别热数据流中的前几个元素（即前缀），并且预取流中元素的平衡（即，后缀）。 由于在运行时注入了热数据流识别码和预取代码，消除了早期预取系统固有的指针相关时间依赖性。 全局确定性有限状态机（DFSM）用于帮助创建用于生成注入到程序中用于前缀检测的代码的概念逻辑。

公开(公告)号：US08108448B2

公开(公告)日：2012-01-31

申请号：US12128684

申请日：2008-05-29

申请人： David K. Siegwart ,  Martin Hirzel

发明人： David K. Siegwart ,  Martin Hirzel

IPC分类号： G06F7/00

CPC分类号： G06F12/0253

摘要： A garbage collection algorithm that achieves hierarchical copy order with parallel garbage collection threads. More specifically, the present invention provides a garbage collection method and system for copying objects from a from-space to a to-space. The method comprises the steps of (a) having multiple threads that simultaneously perform work for garbage collection (GC), (b) examining the placement of objects on blocks, and (c) changing the placement of objects on blocks based on step (b). Preferably, the method includes the additional step of calculating a placement of object(s) based on step (b), and using the result of the calculation for step (c). For example, the calculation may be used to increase the frequency of intra-block pointers and/or to increase the frequency of siblings on the same block.

摘要翻译： 一种垃圾收集算法，可以实现具有并行垃圾收集线程的分层复制顺序。 更具体地说，本发明提供了一种用于从空间到空间复制对象的垃圾收集方法和系统。 该方法包括以下步骤：（a）具有同时执行垃圾回收（GC）的工作的多个线程，（b）检查对象在块上的放置，以及（c）基于步骤（b）改变对象在块上的放置 ）。 优选地，该方法包括基于步骤（b）计算物体的放置并使用步骤（c）的计算结果的附加步骤。 例如，该计算可以用于增加块内指针的频率和/或增加同一块上的兄弟姐妹的频率。

公开(公告)号：US20080235308A1

公开(公告)日：2008-09-25

申请号：US12129067

申请日：2008-05-29

申请人： David K. Siegwart ,  Martin Hirzel

发明人： David K. Siegwart ,  Martin Hirzel

IPC分类号： G06F17/30

CPC分类号： G06F12/0253

摘要： Disclosed is a garbage collection algorithm that achieves hierarchical copy order with parallel garbage collection threads. More specifically, the present invention provides a garbage collection method and system for copying objects from a from-space to a to-space. The method comprises the steps of (a) having multiple threads that simultaneously perform work for garbage collection (GC), (b) examining the placement of objects on blocks, and (c) changing the placement of objects on blocks based on step (b). Preferably, the method includes the additional step of calculating a placement of object(s) based on step (b), and using the result of the calculation for step (c). For example, the calculation may be used to increase the frequency of intra-block pointers and/or to increase the frequency of siblings on the same block.