-
公开(公告)号：US20110071793A1
公开(公告)日：2011-03-24
申请号：US12562659
申请日：2009-09-18
申请人： Evelyn Duesterwald , Peter K. Malkin , Peter F. Sweeney , Qiming Teng , Haichuan Wang , Xiao Zhong
发明人： Evelyn Duesterwald , Peter K. Malkin , Peter F. Sweeney , Qiming Teng , Haichuan Wang , Xiao Zhong
IPC分类号： G06F15/00
CPC分类号： G06F11/3466 , G06F11/3423 , G06F2201/86
摘要： A requesting critical wait time of a given resource may be determined The requesting critical wait time is the time spent by the one or more resources waiting for the given resource, wherein at least one of the resources waiting for the given resource can proceed if access to the given resource is granted. A requested by critical wait time for a resource is determined, the requested by critical wait time being time spent by the resource for waiting solely for the given resource, wherein if the resource were granted access to the given resource, the resource can proceed without further waiting.
摘要翻译： 可以确定给定资源的请求关键等待时间。请求临时等待时间是等待给定资源的一个或多个资源花费的时间，其中等待给定资源的资源中的至少一个可以进行，如果访问 给定的资源被授予。 确定资源的关键等待时间的请求，由关键等待时间所请求的资源用于等待给定资源的时间，其中如果资源被授予对给定资源的访问，则资源可以进一步进行 等候。
-
公开(公告)号：US08521472B2
公开(公告)日：2013-08-27
申请号：US12562659
申请日：2009-09-18
申请人： Evelyn Duesterwald , Peter K. Malkin , Peter F. Sweeney , Qiming Teng , Haichuan Wang , Xiao Zhong
发明人： Evelyn Duesterwald , Peter K. Malkin , Peter F. Sweeney , Qiming Teng , Haichuan Wang , Xiao Zhong
CPC分类号： G06F11/3466 , G06F11/3423 , G06F2201/86
摘要： A requesting critical wait time of a given resource may be determined. The requesting critical wait time is the time spent by the one or more resources waiting for the given resource, wherein at least one of the resources waiting for the given resource can proceed if access to the given resource is granted. A requested by critical wait time for a resource is determined, the requested by critical wait time being time spent by the resource for waiting solely for the given resource, wherein if the resource were granted access to the given resource, the resource can proceed without further waiting.
摘要翻译： 可以确定给定资源的请求临时等待时间。 请求的关键等待时间是等待给定资源的一个或多个资源花费的时间，其中等待给定资源的资源中的至少一个可以在授予对给定资源的访问的情况下继续进行。 确定资源的关键等待时间的请求，由关键等待时间所请求的资源用于等待给定资源的时间，其中如果资源被授予对给定资源的访问，则资源可以进一步进行 等候。
-
3.发明授权Method and system for predicting the performance benefits of mapping subsets of application data to multiple page sizes 有权用于预测将应用程序数据子集映射到多个页面大小的性能优势的方法和系统公开(公告)号：US07376808B2
公开(公告)日：2008-05-20
申请号：US11343565
申请日：2006-01-31
CPC分类号： G06F12/10 , G06F2212/652
摘要： A method for modeling the performance of memory address translation mechanism (MATM), comprises: a) receiving an execution profile that contains a memory address reference stream of an application, a set of page size mappings, and events about the application's data allocations and de-allocations; b) translating each memory reference in the input memory reference stream into a reference to the corresponding data object, by consulting the memory allocation and de-allocation events, to provide a data object reference stream; c) translating each data object reference into a corresponding page reference by consulting the page size mapping and by modeling the data allocation and de-allocation events in accordance with the mapping to provide a page reference stream and a number of pages of each page size that are needed by the respective mapping; d) using the page reference stream to provide a stream of reuse distance values; e) determining, for each reference in the reuse distance value stream, whether the reference results in a hit or a miss reference to the MATM to provide the number of hits and the number of misses for each MATM; f) providing the hit and miss values to a cost model to estimate the number of miss cycles; g) ranking the mappings by their miss cycle values such that the mapping with the lowest number of miss cycles has the highest rank.
摘要翻译： 一种用于对存储器地址转换机制（MATM）的性能进行建模的方法包括：a）接收包含应用的存储器地址参考流，一组页面大小映射和关于应用的数据分配和de的事件的执行简档 分配; b）通过咨询存储器分配和解除分配事件来将输入存储器参考流中的每个存储器引用转换成对相应数据对象的引用，以提供数据对象引用流; c）通过参考所述页面大小映射并且根据所述映射对所述数据分配和解除分配事件进行建模来将每个数据对象引用转换成对应的页面引用，以提供页面引用流和每页页面大小的页数 需要通过各自的映射; d）使用页面参考流来提供重用距离值流; e）针对重用距离值流中的每个参考，确定参考是否导致对MATM的命中或错过引用以提供每个MATM的命中次数和未命中次数; f）向成本模型提供命中和未命中值以估计错过周期数; g）通过它们的错误循环值对映射进行排序，使得具有最低错误周期数的映射具有最高等级。
-
公开(公告)号：US07610266B2
公开(公告)日：2009-10-27
申请号：US11137293
申请日：2005-05-25
IPC分类号： G06F17/30
CPC分类号： G06F11/3466 , Y10S707/99932 , Y10S707/99942
摘要： A method for vertical integrated performance and environment monitoring includes steps, or acts, of: defining one or more events to provide a unified specification; registering one or more events to be detected; detecting an occurrence of at least one of the registered event or events; generating a monitoring entry each time one of the registered events is detected; and entering each of the monitoring entries generated into a single logical entity.
摘要翻译： 用于垂直综合性能和环境监测的方法包括以下步骤或行为：定义一个或多个事件以提供统一的规范; 注册要检测的一个或多个事件; 检测所述注册事件或事件中的至少一个的发生; 每当检测到所登记的事件之一时产生监视条目; 并将生成的每个监视条目输入到单个逻辑实体中。
-
5.发明申请METHOD AND SYSTEM FOR OPTIMIZING COMMUNICATION IN MPI PROGRAMS FOR AN EXECUTION ENVIRONMENT 审中-公开用于优化执行环境MPI程序中的通信的方法和系统公开(公告)号：US20080288957A1
公开(公告)日：2008-11-20
申请号：US11750912
申请日：2007-05-18
申请人： Gheorghe Cascaval , Evelyn Duesterwald , Stephen E. Smith , Peter F. Sweeney , Robert W. Wisniewski
发明人： Gheorghe Cascaval , Evelyn Duesterwald , Stephen E. Smith , Peter F. Sweeney , Robert W. Wisniewski
IPC分类号： G06F13/00
CPC分类号： G06F9/5038 , G06F2209/508
摘要： A system and method for mapping application tasks to processors in a computing environment that takes into account the hardware communication topology of a machine and an application communication pattern. The hardware communication topology (HCT) is defined according to hardware parameters affecting communication between two tasks, such as connectivity, bandwidth and latency; and, the application communication pattern (ACP) is defined to mean the number and size of bytes that are communicated between the different pairs of communicating tasks. By collecting information on the messages exchanged by the tasks that communicate, the communication pattern of the application may be determined. By combing the HCT and ACP a cost model for a given mapping can be determined. Any algorithm computing a mapping can use the HCT, ACP, and the cost model, thus the combination of an HCT, ACP, and cost model allow an automatically optimized mapping of tasks to processing elements to be achieved
摘要翻译： 一种将应用任务映射到计算环境中的处理器的系统和方法，其考虑了机器的硬件通信拓扑和应用通信模式。 硬件通信拓扑（HCT）根据影响两个任务之间通信的硬件参数进行定义，如连接性，带宽和延迟; 并且应用通信模式（ACP）被定义为在不同的通信任务对之间通信的字节的数量和大小。 通过收集由通信任务交换的消息的信息，可以确定应用的通信模式。 通过组合HCT和ACP，可以确定给定映射的成本模型。 计算映射的任何算法都可以使用HCT，ACP和成本模型，因此HCT，ACP和成本模型的组合允许任务自动优化到要实现的处理元素的映射
-
公开(公告)号：US08527624B2
公开(公告)日：2013-09-03
申请号：US12130718
申请日：2008-05-30
申请人： Ying Chen , Evelyn Duesterwald , Ying Li , Qi Ming Teng
发明人： Ying Chen , Evelyn Duesterwald , Ying Li , Qi Ming Teng
IPC分类号： G06F15/173 , G06F11/00 , G06F9/44 , G01R11/00 , G01R31/28
CPC分类号： G06F11/3612 , G06F11/3452 , G06F11/3466 , G06F2201/86 , G06F2201/865 , G06F2201/88
摘要： Adaptive profiling for performance analysis of a computer system controls one or more agents to monitor a plurality of events occurring in a target computer system based on an adaptive logic. Collected data may be filtered and analyzed to determine one or more contributor events that attribute to performance of the target computer system. One or more patterns are observed or detected in said collected data, behavior of said one or more agents are adjusted based on said detected one or more patterns. The adaptive logic may be further reconfigured based on said detected one or more patterns.
摘要翻译： 用于计算机系统的性能分析的自适应分析控制一个或多个代理以基于自适应逻辑监视在目标计算机系统中发生的多个事件。 收集的数据可以被过滤和分析以确定属于目标计算机系统的性能的一个或多个贡献者事件。 在所述收集的数据中观察或检测到一个或多个图案，基于所检测到的一个或多个图案来调整所述一个或多个代理的行为。 可以基于所检测到的一个或多个模式来进一步重新配置自适应逻辑。
-
7.发明申请公开(公告)号：US20110209119A1
公开(公告)日：2011-08-25
申请号：US12709712
申请日：2010-02-22
IPC分类号： G06F9/44
CPC分类号： G06F8/72
摘要： Interactive iterative program parallelization based on dynamic feedback program parallelization, in one aspect, may identify a ranked list of one or more candidate pieces of code each with one or more source refactorings that can be applied to parallelize the code, apply at least one of the one or more refactorings to create a revised code, and determine performance data associated with the revised code. The performance data may be used to make decisions on identifying next possible ranked list of refactorings.
摘要翻译： 在一个方面，基于动态反馈程序并行化的交互式迭代程序并行化可以识别一个或多个候选代码片段的排序列表，每个代码片段具有可以应用于并行化代码的一个或多个源重构，应用至少一个 一个或多个重构以创建经修改的代码，并且确定与修改的代码相关联的性能数据。 性能数据可以用于对识别下一个可能排列的重构列表做出决定。
-
8.发明申请Mechanism for on-line prediction of future performance measurements in a computer system 有权在线预测计算机系统中未来性能测量的机制公开(公告)号：US20060217940A1
公开(公告)日：2006-09-28
申请号：US11417828
申请日：2006-05-04
IPC分类号： G06F15/00
CPC分类号： G06F11/008 , G06F1/206 , G06F1/3203 , G06F1/329 , G06F9/4881 , G06F11/3419 , G06F11/3447 , G06F11/3452 , G06F2201/88 , G06F2209/5019 , Y02D10/24
摘要： Disclosed are a method and system for predicting future values of a target metric associated with a task executed on a computer system. The method comprises the steps of, over a given period of time, measuring at least one defined metric, transforming that measurement into a value for a predictor source metric, and using the value for the predictor source metric to obtain a predicted future value for said target metric. The preferred embodiment of this invention provides a flexible performance multi-predictor to solve the problem of providing accurate future behavior predictions for adaptive reconfiguration systems. The multi-predictor makes predictions about future workload characteristic by periodically reading available hardware counters. Also disclosed is a method and system for periodically reconfiguring an adaptive computer system by rescheduling tasks based on future behavior predictions.
摘要翻译： 公开了一种用于预测与在计算机系统上执行的任务相关联的目标度量的未来值的方法和系统。 该方法包括以下步骤：在给定的时间段内，测量至少一个定义的度量，将该测量值变换为预测器源度量的值，以及使用预测值源度量的值来获得所述预测器源度量的预测未来值 目标度量。 本发明的优选实施例提供了一种灵活的性能多预测器来解决为自适应重配置系统提供准确的未来行为预测的问题。 多预测器通过定期读取可用的硬件计数器来预测未来的工作负载特征。 还公开了一种通过基于未来行为预测重新安排任务来周期性重新配置自适应计算机系统的方法和系统。
-
9.发明授权Method and system for fast unlinking of a linked branch in a caching dynamic translator 失效在缓存动态翻译器中快速取消链接分支的方法和系统公开(公告)号：US06725335B2
公开(公告)日：2004-04-20
申请号：US09755780
申请日：2001-01-05
IPC分类号： G06F1200
CPC分类号： G06F9/30174 , G06F9/44521 , G06F9/45504 , G06F11/3471 , G06F11/3476 , G06F2201/88 , G06F2201/885
摘要： In a system and method for linking and unlinking code fragments stored in a code cache, a memory area is associated with a branch in a first code fragment that branches outside the cache. If the branch can be set to branch to a location in a second code fragment stored in the cache, branch reconstruction information is stored in the memory area associated with the branch, and the branch instruction is updated to branch to the location in the second code fragment, thereby linking the first code fragment to the second code fragment. If it is determined that the previously linked branch should be unlinked, the first and second code fragments at that branch are unlinked by reading the information stored in the associated memory area at the time of linking, and using that information to reset the branch to its state prior to the linking.
摘要翻译： 在用于链接和解链存储在代码高速缓存中的代码片段的系统和方法中，存储器区域与分支在高速缓存之外的第一代码片段中的分支相关联。 如果可以将分支设置为分支到存储在高速缓存中的第二代码片段中的位置，则分支重建信息被存储在与分支相关联的存储区域中，并且分支指令被更新以分支到第二代码中的位置 片段，从而将第一代码片段链接到第二代码片段。 如果确定先前链接的分支应该被取消链接，则在该分支处的第一和第二代码片段在链接时读取存储在相关联的存储器区域中的信息而被取消链接，并且使用该信息将分支重置为其 链接之前的状态。
-
10.发明申请Mechanism for on-line prediction of future performance measurements in a computer system 失效在线预测计算机系统中未来性能测量的机制公开(公告)号：US20050086029A1
公开(公告)日：2005-04-21
申请号：US10688293
申请日：2003-10-17
CPC分类号： G06F11/008 , G06F1/206 , G06F1/3203 , G06F1/329 , G06F9/4881 , G06F11/3419 , G06F11/3447 , G06F11/3452 , G06F2201/88 , G06F2209/5019 , Y02D10/24
摘要： Disclosed are a method and system for predicting future values of a target metric associated with a task executed on a computer system. The method comprises the steps of, over a given period of time, measuring at least one defined metric, transforming that measurement into a value for a predictor source metric, and using the value for the predictor source metric to obtain a predicted future value for said target metric. The preferred embodiment of this invention provides a flexible performance multi-predictor to solve the problem of providing accurate future behavior predictions for adaptive reconfiguration systems. The multi-predictor makes predictions about future workload characteristic by periodically reading available hardware counters. Also disclosed is a method and system for periodically reconfiguring an adaptive computer system by rescheduling tasks based on future behavior predictions.
摘要翻译： 公开了一种用于预测与在计算机系统上执行的任务相关联的目标度量的未来值的方法和系统。 该方法包括以下步骤：在给定的时间段内，测量至少一个定义的度量，将该测量值变换为预测器源度量的值，以及使用预测值源度量的值来获得所述预测器源度量的预测未来值 目标度量。 本发明的优选实施例提供了一种灵活的性能多预测器来解决为自适应重配置系统提供准确的未来行为预测的问题。 多预测器通过定期读取可用的硬件计数器来预测未来的工作负载特征。 还公开了一种通过基于未来行为预测重新安排任务来周期性重新配置自适应计算机系统的方法和系统。
-
-
-
-
-
-
-
-
-
搜索结果
28 条记录 (耗时 0.038 秒)
