-
公开(公告)号：US06381740B1
公开(公告)日：2002-04-30
申请号：US08931151
申请日：1997-09-16
IPC分类号： G06F900
CPC分类号： G06F8/445
摘要： A method and system for incrementally improving the layout of a program image of a computer program to reduce the working set. The system iteratively selects pairs of basic blocks and reorders the basic blocks in the range delimited by the selected pair of basic blocks. The system selects the pairs of basic blocks so that the working set of the computer program is improved by reordering the basic block in the range. Thus, during each iteration, the working set is improved. The system continues with these iterations until a termination condition (e.g., number of iterations) is satisfied. In one embodiment, during each iteration the system designates one of the basic blocks as an initial anchor basic block. The system then repeats the following until the same range of basic blocks is identified twice in a row. The system first finds a basic block such that when the basic blocks in the range from the anchor basic block to the found basic block are reordered, the working set is more favorable than reordering any other range that ends with the anchor basic block. The system then designates the found basic block as the new anchor basic block. When the same range is found twice in a row, the system reorders the basic blocks in the range. This process is repeated for each iteration until a termination condition is satisfied. The resulting reordered program image has its working set improved.
摘要翻译： 一种用于逐步改进计算机程序的程序图像的布局以减少工作集合的方法和系统。 系统迭代地选择一对基本块，并且在由所选择的一对基本块分隔的范围内重新排序基本块。 系统选择基本块对，通过重新排序该范围内的基本块来提高计算机程序的工作集。 因此，在每次迭代期间，改进了工作集。 系统继续进行这些迭代，直到满足终止条件（例如，迭代次数）为止。 在一个实施例中，在每次迭代期间，系统将基本块中的一个指定为初始锚基本块。 然后，系统重复以下操作，直到相同范围的基本块被连续两次识别。 系统首先找到一个基本块，使得当从锚基本块到找到的基本块的范围内的基本块被重新排序时，工作集比重新排序以锚基本块结束的任何其他范围更有利。 然后，系统将找到的基本块指定为新的锚基本块。 当连续两次找到相同的范围时，系统重新排列范围内的基本块。 对于每个迭代重复该过程，直到满足终止条件。 所产生的重新排序的程序图像的工作集改进了。
-
公开(公告)号：US06658648B1
公开(公告)日：2003-12-02
申请号：US08931861
申请日：1997-09-16
IPC分类号： G06F944
CPC分类号： G06F9/44557 , G06F9/445
摘要： A method and system for improving the working set of a program image. The working set (WS) improvement system of the present invention employs a two-phase technique for improving the working set. In the first phase, the WS improvement system inputs the program image and outputs a program image with the locality of its references improved. In the second phase, the WS improvement system inputs the program image with its locality of references improved and outputs a program image with the placement of its basic blocks in relation to page boundaries improved so that the working set is reduced.
摘要翻译： 一种改善程序图像工作集的方法和系统。 本发明的工作组（WS）改进系统采用两相技术改进工作组。 在第一阶段，WS改进系统输入程序图像，并输出其参考文件的位置改进了的程序图像。 在第二阶段，WS改进系统输入节目图像，其参考地点改善，并且输出节目图像，其基本块相对于页面边界的放置得到改善，从而减少了工作集。
-
公开(公告)号：US07958176B2
公开(公告)日：2011-06-07
申请号：US11407639
申请日：2006-04-20
IPC分类号： G06F17/10
CPC分类号： G06F9/44557 , G06F9/445
摘要： Techniques for identifying coefficients are described. A technique collects sample input values to the filter and identifies desired output values from the filter for the collected sample input values. The technique then generates a power spectrum of the collected sample input values and a power spectrum of the identified desired output values. The technique then calculates the difference between the generated power spectra. Finally, the technique identifies coefficients that yield a filter transfer function that closely approximates the calculated differences. A technique identifies coefficients for a finite impulse response filter. The technique collects sample input values for a function and identifies desired output values for the filter for the collected sample input values. The technique then approximates the output values from the input values using a linear fitting technique. Finally, the technique sets the coefficients to values obtained from the linear-fitting technique.
摘要翻译： 描述用于识别系数的技术。 技术将采样输入值收集到滤波器，并从滤波器识别收集的采样输入值的所需输出值。 然后，该技术产生所收集的采样输入值的功率谱和所识别的期望输出值的功率谱。 然后，该技术计算所产生的功率谱之间的差异。 最后，该技术识别产生近似于计算出的差异的滤波器传递函数的系数。 一种技术识别有限脉冲响应滤波器的系数。 该技术收集函数的样本输入值，并为收集的样本输入值识别滤波器的所需输出值。 然后，该技术使用线性拟合技术从输入值近似输出值。 最后，该技术将系数设置为从线性拟合技术获得的值。
-
公开(公告)号：US5903917A
公开(公告)日：1999-05-11
申请号：US797589
申请日：1997-02-07
CPC分类号： G06F12/10 , G06F12/0864
摘要： A computer system and method for aligning blocks within source program image in accordance with alignment constraints. The source program image is divided in pages. The alignment system receives an indication of which of the blocks are accessed during each time interval while the source program image is executed. The alignment system determines from the received indications which of the pages are accessed during each time interval. The alignment system then identifies the first block of the source program image that has an alignment constraint and repeats the following until all blocks have been aligned. The alignment system identifies a group of contiguous blocks that are adjacent to the identified block such that the identified blocks can be rearranged without increasing the number of pages that are accessed during each time interval. The alignment system then determines an arrangement of the blocks that tends to minimize the amount of padding that is needed to satisfy the alignment constraints. The alignment system then stores the rearranged blocks along with the padding in a destination program image.
摘要翻译： 一种用于根据对准约束在源程序图像内对准块的计算机系统和方法。 源程序图像分为页面。 在执行源程序图像时，对准系统接收在每个时间间隔期间访问哪个块的指示。 对准系统从接收到的指示确定在每个时间间隔期间访问哪些页面。 对准系统然后识别具有对准约束的源程序图像的第一块，并重复以下操作，直到所有块都对齐。 对准系统识别与所识别的块相邻的一组连续的块，使得可以重新排列所识别的块，而不增加在每个时间间隔期间被访问的页数。 对准系统然后确定倾向于最小化满足对准约束所需的填充量的块的布置。 对准系统然后将重排的块与填充一起存储在目的节目图像中。
-
公开(公告)号：US07181736B2
公开(公告)日：2007-02-20
申请号：US10668764
申请日：2003-09-23
IPC分类号： G06F9/45
CPC分类号： G06F9/44557 , G06F9/445
摘要： A method and system for improving the working set of a program image. The working set (WS) improvement system of the present invention employs a two-phase technique for improving the working set. In the first phase, the WS improvement system inputs the program image and outputs a program image with the locality of its references improved. In the second phase, the WS improvement system inputs the program image with its locality of references improved and outputs a program image with the placement of its basic blocks in relation to page boundaries improved so that the working set is reduced.
摘要翻译： 一种改善程序图像工作集的方法和系统。 本发明的工作组（WS）改进系统采用两相技术改进工作组。 在第一阶段，WS改进系统输入程序图像，并输出其参考文件的位置改进了的程序图像。 在第二阶段，WS改进系统输入节目图像，其参考地点改善，并且输出节目图像，其基本块相对于页面边界的放置得到改善，从而减少了工作集。
-
6.发明授权Method and system for improving the layout of a program image using clustering 失效使用聚类改进程序图像布局的方法和系统公开(公告)号：US06269477B1
公开(公告)日：2001-07-31
申请号：US08931030
申请日：1997-09-16
IPC分类号： G06F945
CPC分类号： G06F8/4442
摘要： A computer system for establishing a program image layout of a computer program. The computer program has basic blocks that each have temporal usage vector indicating time intervals during which the basic block was accessed during an execution of the computer program. For each of the basic blocks, the layout system initializes a cluster to contain that basic block and the layout system repeats the following until one cluster remains that contains all of the basic blocks. For each possible ordered pair of clusters, the layout system determines a metric value for the ordered pair of clusters, the metric value being derived from the temporal usage vectors of the basic blocks of the cluster and estimating an effect on the performance of the computer program when that ordered pair of clusters are position in the program image in that order and contiguously. The layout system then selects the ordered pair whose metric value estimates the smallest effect on the performance of the computer program. The layout system then replaces the clusters of the selected ordered pair of clusters with a new cluster that includes the basic blocks of the selected order pair.
摘要翻译： 一种用于建立计算机程序的程序图像布局的计算机系统。 计算机程序具有基本块，每个基本块具有指示在执行计算机程序期间访问基本块的时间间隔的时间使用向量。 对于每个基本块，布局系统初始化集群以包含该基本块，并且布局系统重复以下，直到包含所有基本块的一个集群为止。 对于每个可能有序的聚类对，布局系统确定有序聚类对的度量值，度量值是从群集的基本块的时间使用向量导出的，并且估计对计算机程序的性能的影响 当有序对的簇在程序图像中以该顺序并且连续地位置时。 然后，布局系统选择其度量值估计对计算机程序的性能的最小影响的有序对。 然后，布局系统用包括所选订单对的基本块的新集群替换所选择的有序对集群的集群。
-
7.发明授权公开(公告)号：US08990538B2
公开(公告)日：2015-03-24
申请号：US12940780
申请日：2010-11-05
CPC分类号： G06F11/3037 , G06F11/3055 , G06F12/0223 , G06F12/06 , G06F2212/205 , G11C16/349
摘要： A method and a memory manager for managing data storage in a plurality of types of memories. The types of memories may comprise a primary memory, such as DRAM, and a secondary memory, such as a phase change memory (PCM) or Flash memory, which may have a limited lifetime. The memory manager may be part of an operating system and may manage the memories as part of a unified address space. Characteristics of data to be stored in the memories may be used to select between the primary and secondary memories to store the data and move data between the memories. When the data is to be stored in the secondary memory, health information on the secondary memory and characteristics of the data to be stored may be used to select a location within the secondary memory to store the data.
摘要翻译： 一种用于管理多种类型的存储器中的数据存储的方法和存储器管理器。 存储器的类型可以包括诸如DRAM的主存储器，以及可能具有有限寿命的次要存储器，例如相变存储器（PCM）或闪速存储器。 存储器管理器可以是操作系统的一部分，并且可以作为统一地址空间的一部分来管理存储器。 要存储在存储器中的数据的特征可以用于在主存储器和次存储器之间选择存储数据并在存储器之间移动数据。 当数据要存储在辅助存储器中时，辅助存储器上的健康信息和要存储的数据的特性可以用于选择次存储器内的存储数据的位置。
-
公开(公告)号：US07373418B2
公开(公告)日：2008-05-13
申请号：US10931074
申请日：2004-08-31
申请人： William J. Bolosky , Craig M Dowell , Robert P. Fitzgerald , Steven P. Levi , Jan de Rie , Richard F. Rashid
发明人： William J. Bolosky , Craig M Dowell , Robert P. Fitzgerald , Steven P. Levi , Jan de Rie , Richard F. Rashid
IPC分类号： G06F15/16
CPC分类号： H04L65/4084 , G06F19/00 , H04L29/06 , H04L29/06027 , H04L65/1069 , H04L65/4092 , H04L69/14
摘要： A wire protocol provides message formats for creating multiple network connections between a media server and a client. These multiple network connections may include a control link connection for passing control information and a data funnel connection for passing data of multiple media. The data funnel connection may be a multipoint-to-point connection that connects multiple data servers with the client. The protocol facilitates multiple requests being concurrently outstanding and asynchronous processing of requests. The protocol is designed to exist on top of a transport protocol layer.
-
9.发明授权Method and system for scheduling the use of a computer system resource using a resource planner and a resource provider 失效使用资源规划者和资源提供者调度计算机系统资源使用的方法和系统
公开(公告)号：US6003061A
公开(公告)日：1999-12-14
申请号：US569048
申请日：1995-12-07
申请人： Michael B. Jones , Paul J. Leach , Richard P. Draves, Jr. , Joseph S. Barrera, III , Steven P. Levi , Richard F. Rashid , Robert P. Fitzgerald
发明人： Michael B. Jones , Paul J. Leach , Richard P. Draves, Jr. , Joseph S. Barrera, III , Steven P. Levi , Richard F. Rashid , Robert P. Fitzgerald
CPC分类号： G06F9/50 , G06F2209/5014
摘要： A method and system for scheduling the use of a computer system resource using a resource planner and a resource provider is provided. In a preferred embodiment, a resource is scheduled for use by a plurality of consumer entities. Each consumer entity may request the commitment of a share of the resource. The method and system utilizes representations of resource usage policy, present commitments of shares of the resource, and present commitments of specified amounts of the resource over specified period of time. The method and system first receives a request from a consumer entity for the commitment of a specified share of the resource. In response, the method and system determines whether the specified share of the resource should be committed to the requesting consumer entity. This determination is based on the representations of resource usage policy and present commitments of shares of the resource. If it is determined that the specified share of the resource should be committed to the requesting consumer entity, then the method and system modifies the representation of present commitments of shares of the resource to commit the specified share of the resource to the requesting consumer entity. The method and system then schedules the use of the resource by the plurality of consumer entity based on the modified representation of present commitments of shares of the resource.
摘要翻译： 提供了一种用于使用资源规划器和资源提供者来调度计算机系统资源的使用的方法和系统。 在优选实施例中，资源被调度为由多个消费者实体使用。 每个消费者实体可以请求资源份额的承诺。 该方法和系统利用资源使用政策的表示，资源份额的现有承诺以及指定时间内资源规定数量的承诺。 方法和系统首先从消费者实体接收对资源的指定份额的承诺的请求。 作为响应，方法和系统确定资源的指定份额是否应该提交给请求的消费者实体。 这一决定是基于资源使用政策的表示和资源份额的现有承诺。 如果确定资源的指定份额应该提交给请求的消费者实体，那么该方法和系统将修改资源共享的当前承诺的表示，以将该资源的指定份额提交给请求的消费者实体。 该方法和系统随后基于对资源份额的当前承诺的修改表示来安排多个消费者实体对资源的使用。
-
10.发明授权Method and system for scheduling the execution of threads using optional time-specific scheduling constraints 失效使用可选的时间特定调度约束调度线程执行的方法和系统
公开(公告)号：US5812844A
公开(公告)日：1998-09-22
申请号：US568577
申请日：1995-12-07
申请人： Michael B. Jones , Paul J. Leach , Richard P. Draves, Jr. , Joseph S. Barrera, III , Steven P. Levi , Richard F. Rashid , Robert P. Fitzgerald
发明人： Michael B. Jones , Paul J. Leach , Richard P. Draves, Jr. , Joseph S. Barrera, III , Steven P. Levi , Richard F. Rashid , Robert P. Fitzgerald
CPC分类号： G06F9/4887
摘要： A method and system for scheduling the execution of threads using optional time-specific constraints is provided. In a preferred embodiment, a scheduler schedules the execution of a plurality of threads each specifying either a percentage processor time scheduling constraint, indicating that a certain percentage of processor time should be dedicated to the execution of the thread, or a deadline scheduling constraint, indicating that a specified quantity of work should be done by the thread by a specified time deadline. For each thread specifying a percentage processor time scheduling constraint, the scheduler determines a restart time corresponding to the percentage processor time scheduling constraint. For each thread specifying a deadline scheduling constraint, the scheduler determines a restart time corresponding to the deadline scheduling constraint. The scheduler then utilizes the determined restart times for scheduling the threads by selecting the thread having the earliest restart time for execution.
摘要翻译： 提供了一种使用可选的时间特定约束调度线程执行的方法和系统。 在优选实施例中，调度器调度多个线程的执行，每个线程指定百分比处理器时间调度约束，指示处理器时间的一定百分比应该专用于线程的执行，或者指示最后期限调度约束 线程应在指定的时间期限内完成指定数量的工作。 对于指定百分比处理器时间调度约束的每个线程，调度器确定对应于百分比处理器时间调度约束的重新启动时间。 对于指定期限调度约束的每个线程，调度器确定对应于期限调度约束的重启时间。 调度器然后利用所确定的重新启动次数来调度线程，方法是选择具有最早执行重启时间的线程。
-
-
-
-
-
-
-
-
-
搜索结果
28 条记录 (耗时 0.03 秒)
