公开(公告)号：US08539455B2

公开(公告)日：2013-09-17

申请号：US12055996

申请日：2008-03-26

申请人： Erik Berg ,  Erik Hagersten ,  Hakan Zeffer ,  Magnus Vesterlund ,  Mats Nilsson ,  Mikael Petterson

发明人： Erik Berg ,  Erik Hagersten ,  Hakan Zeffer ,  Magnus Vesterlund ,  Mats Nilsson ,  Mikael Petterson

IPC分类号： G06F9/44 ,  G06F15/00 ,  G06F12/00

CPC分类号： G06F11/3447 ,  G06F11/3466 ,  G06F2201/88 ,  G06F2201/885

摘要： A system, method, and computer program product that captures performance-characteristic data from the execution of a program and models system performance based on that data. Performance-characterization data based on easily captured reuse distance metrics is targeted. Reuse distance for one memory operation may be measured as the number of memory operations that have been performed since the memory object it accesses was last accessed. Separate call stacks leading up to the same memory operation are identified and statistics are separated for the different call stacks. Methods for efficiently capturing this kind of metrics are described. These data can be refined into easily interpreted performance metrics, such as performance data related to caches with LRU replacement and random replacement strategies in combination with fully associative as well as limited associativity cache organizations. Methods for assessing cache utilization as well as parallel execution are covered. The method includes modeling multithreaded memory systems and detecting false sharing coherence misses.

摘要翻译： 一种系统，方法和计算机程序产品，可从执行程序中捕获性能特征数据，并根据该数据对系统性能进行建模。 基于容易捕获的重用距离度量的性能表征数据是针对性的。 一个存储器操作的重复使用距离可以被测量为自上次访问其访问的存储器对象以来已执行的存储器操作的数量。 识别导致相同内存操作的单独的调用堆栈，并为不同的调用堆栈分隔统计信息。 描述有效捕获这种度量的方法。 这些数据可以被简化为易于解释的性能指标，例如与具有LRU替换的缓存和随机替换策略相关联的性能数据以及完全关联以及有限关联缓存组织。 涵盖了评估缓存利用率和并行执行的方法。 该方法包括建模多线程内存系统并检测虚假共享连贯性错误。

公开(公告)号：US20090055594A1

公开(公告)日：2009-02-26

申请号：US11806997

申请日：2007-06-05

申请人： Erik Berg ,  Erik Hagersten ,  Mats Nilsson ,  Mikael Petterson ,  Magnus Vesterlund ,  Hakan Zeffer

发明人： Erik Berg ,  Erik Hagersten ,  Mats Nilsson ,  Mikael Petterson ,  Magnus Vesterlund ,  Hakan Zeffer

IPC分类号： G06F12/12

CPC分类号： G06F11/3644 ,  G06F11/3447 ,  G06F11/3466 ,  G06F2201/88 ,  G06F2201/885

摘要： A system for, method of and computer program product captures performance-characteristic data from the execution of a program and models system performance based on that data. Performance-characterization data based on easily captured reuse distance metrics is targeted, defined as the total number of memory references between two accesses to the same piece of data. Methods for efficiently capturing this kind of metrics are described. These data can be refined into easily interpreted performance metrics, such as performance data related to caches with LRU replacement and random replacement strategies in combination with fully associative as well as limited associativity cache organizations.

摘要翻译： 系统，方法和计算机程序产品从执行程序中捕获性能特征数据，并根据该数据对系统性能进行建模。 基于容易捕获的复用距离度量的性能表征数据被定义为两次访问同一数据块之间的存储器引用的总数。 描述有效捕获这种度量的方法。 这些数据可以被简化为易于解释的性能指标，例如与具有LRU替换的缓存和随机替换策略相关联的性能数据以及完全关联以及有限关联缓存组织。

公开(公告)号：US08443341B2

公开(公告)日：2013-05-14

申请号：US11938247

申请日：2007-11-09

申请人： Erik Berg ,  Erik Hagersten ,  Mats Nilsson ,  Mikael Petterson ,  Magnus Vesterlund ,  Hakan Zeffer

发明人： Erik Berg ,  Erik Hagersten ,  Mats Nilsson ,  Mikael Petterson ,  Magnus Vesterlund ,  Hakan Zeffer

IPC分类号： G06F9/44

CPC分类号： G06F11/3616 ,  G06F12/0802

摘要： A system for, method of and computer program product captures performance-characteristic data from the execution of a program and models system performance based on that data. Performance-characterization data based on easily captured reuse distance metrics is targeted, defined as the total number of memory references between two accesses to the same piece of data. Methods for efficiently capturing this kind of metrics are described. These data can be refined into easily interpreted performance metrics, such as performance data related to caches with LRU replacement and random replacement strategies in combination with fully associative as well as limited associativity cache organizations. Methods for assessing cache utilization as well as parallel execution are covered.

摘要翻译： 系统，方法和计算机程序产品从执行程序中捕获性能特征数据，并根据该数据对系统性能进行建模。 基于容易捕获的复用距离度量的性能表征数据被定义为两次访问同一数据块之间的存储器引用的总数。 描述有效捕获这种度量的方法。 这些数据可以被简化为易于解释的性能指标，例如与具有LRU替换的缓存和随机替换策略相关联的性能数据以及完全关联以及有限关联缓存组织。 涵盖了评估缓存利用率和并行执行的方法。

公开(公告)号：US20080244533A1

公开(公告)日：2008-10-02

申请号：US12055996

申请日：2008-03-26

申请人： Erik Berg ,  Erik Hagersten ,  Hakan Zeffer ,  Magnus Vesterlund ,  Mats Nilsson ,  Mikael Petterson

发明人： Erik Berg ,  Erik Hagersten ,  Hakan Zeffer ,  Magnus Vesterlund ,  Mats Nilsson ,  Mikael Petterson

IPC分类号： G06F9/44

CPC分类号： G06F11/3447 ,  G06F11/3466 ,  G06F2201/88 ,  G06F2201/885

摘要： A system for, method of and computer program product captures performance-characteristic data from the execution of a program and models system performance based on that data. Performance-characterization data based on easily captured reuse distance metrics is targeted, defined as the total number of memory references between two accesses to the same piece of data. Methods for efficiently capturing this kind of metrics are described. These data can be refined into easily interpreted performance metrics, such as performance data related to caches with LRU replacement and random replacement strategies in combination with fully associative as well as limited associativity cache organizations. Methods for assessing cache utilization as well as parallel execution are covered.

摘要翻译： 系统，方法和计算机程序产品从执行程序中捕获性能特征数据，并根据该数据对系统性能进行建模。 基于容易捕获的复用距离度量的性能表征数据被定义为两次访问同一数据块之间的存储器引用的总数。 描述有效捕获这种度量的方法。 这些数据可以被简化为易于解释的性能指标，例如与具有LRU替换的缓存和随机替换策略相关联的性能数据以及完全关联以及有限关联缓存组织。 涵盖了评估缓存利用率和并行执行的方法。

公开(公告)号：US08141058B2

公开(公告)日：2012-03-20

申请号：US11806997

申请日：2007-06-05

申请人： Erik Berg ,  Erik Hagersten ,  Mats Nilsson ,  Mikael Petterson ,  Magnus Vesterlund ,  Hakan Zeffer

发明人： Erik Berg ,  Erik Hagersten ,  Mats Nilsson ,  Mikael Petterson ,  Magnus Vesterlund ,  Hakan Zeffer

IPC分类号： G06F9/44 ,  G06F9/45

CPC分类号： G06F11/3644 ,  G06F11/3447 ,  G06F11/3466 ,  G06F2201/88 ,  G06F2201/885

摘要： A system for, method of and computer program product captures performance-characteristic data from the execution of a program and models system performance based on that data. Performance-characterization data based on easily captured reuse distance metrics is targeted, defined as the total number of memory references between two accesses to the same piece of data. Methods for efficiently capturing this kind of metrics are described. These data can be refined into easily interpreted performance metrics, such as performance data related to caches with LRU replacement and random replacement strategies in combination with fully associative as well as limited associativity cache organizations.

摘要翻译： 系统，方法和计算机程序产品从执行程序中捕获性能特征数据，并根据该数据对系统性能进行建模。 基于容易捕获的复用距离度量的性能表征数据被定义为两次访问同一数据块之间的存储器引用的总数。 描述有效捕获这种度量的方法。 这些数据可以被简化为易于解释的性能指标，例如与具有LRU替换的缓存和随机替换策略相关联的性能数据以及完全关联以及有限关联缓存组织。

公开(公告)号：US20090125465A1

公开(公告)日：2009-05-14

申请号：US11938247

申请日：2007-11-09

申请人： Erik Berg ,  Erik Hagersten ,  Mats Nilsson ,  Mikael Petterson ,  Magnus Vesterlund ,  Hakan Zeffer

发明人： Erik Berg ,  Erik Hagersten ,  Mats Nilsson ,  Mikael Petterson ,  Magnus Vesterlund ,  Hakan Zeffer

IPC分类号： G06F15/18

CPC分类号： G06F11/3616 ,  G06F12/0802

摘要： A system for, method of and computer program product captures performance-characteristic data from the execution of a program and models system performance based on that data. Performance-characterization data based on easily captured reuse distance metrics is targeted, defined as the total number of memory references between two accesses to the same piece of data. Methods for efficiently capturing this kind of metrics are described. These data can be refined into easily interpreted performance metrics, such as performance data related to caches with LRU replacement and random replacement strategies in combination with fully associative as well as limited associativity cache organizations. Methods for assessing cache utilization as well as parallel execution are covered.

摘要翻译： 系统，方法和计算机程序产品从执行程序中捕获性能特征数据，并根据该数据对系统性能进行建模。 基于容易捕获的复用距离度量的性能表征数据被定义为两次访问同一数据块之间的存储器引用的总数。 描述有效捕获这种度量的方法。 这些数据可以被简化为易于解释的性能指标，例如与具有LRU替换的缓存和随机替换策略相关联的性能数据以及完全关联以及有限关联缓存组织。 涵盖了评估缓存利用率和并行执行的方法。

公开(公告)号：US08170859B1

公开(公告)日：2012-05-01

申请号：US11412831

申请日：2006-04-28

申请人： Magnus Christensson ,  Samuel Rydh ,  Magnus Vesterlund ,  Johan Rydberg

发明人： Magnus Christensson ,  Samuel Rydh ,  Magnus Vesterlund ,  Johan Rydberg

IPC分类号： G06F9/45 ,  G06F9/455

CPC分类号： G06F11/3457 ,  G06F9/455 ,  G06F9/4812 ,  G06F11/3419 ,  G06F11/3616 ,  G06F2201/81 ,  G06F2201/815 ,  G06F2201/865 ,  G06F2201/88

摘要： Arbitrary, unmodified code and/or software may be executed directly on a host processor operating in a virtualized mode using hardware virtualization support and performance counters. The arbitrary software may be run on the host processor until the host processor exits from the virtualized mode. An end execution time may be calculated in response to the host processor exiting from the virtualized mode. An event may then be handled based on an execution time at which the host processor exited from the virtualized mode and a time at which a scheduled event was to occur.

摘要翻译： 可以使用硬件虚拟化支持和性能计数器直接在使用虚拟化模式运行的主机处理器上执行任意的未修改的代码和/或软件。 任意软件可以在主机处理器上运行，直到主机处理器退出虚拟化模式。 可以响应于主机处理器从虚拟化模式退出来计算结束执行时间。 然后可以基于主机处理器从虚拟化模式退出的执行时间和预定事件发生的时间来处理事件。