公开(公告)号：US09389992B2

公开(公告)日：2016-07-12

申请号：US14820798

申请日：2015-08-07

Applicant: Microsoft Technology Licensing, LLC

Inventor： Renat Gataullin ,  Alexander G. Gounares ,  Christopher W. Fraser

IPC: G06F9/44 ,  G06F11/36 ,  G06F11/34 ,  G06F11/32

CPC classification number: G06F11/3688 ,  G06F8/70 ,  G06F11/3003 ,  G06F11/3082 ,  G06F11/3093 ,  G06F11/323 ,  G06F11/3466 ,  G06F11/3636 ,  G06F2201/865

Abstract: A tracing system may use different configurations for tracing various functions in different manners. A configuration may be a group of settings that may define which data elements to collect, as well as the manner in which the data may be summarized, stored, and in some cases, displayed. Example configurations may include debugging configuration, performance optimization configuration, long term monitoring configuration, and others. The tracing system may be able to trace one group of functions with one configuration, while tracing another group of functions in the same application using a different configuration.

Abstract translation: 跟踪系统可以使用不同的配置来以不同的方式跟踪各种功能。 配置可以是可以定义要收集哪些数据元素的一组设置，以及可以总结，存储和在某些情况下显示数据的方式。 示例配置可能包括调试配置，性能优化配置，长期监视配置等。 跟踪系统可能能够使用一种配置跟踪一组功能，同时使用不同的配置跟踪同一应用程序中的另一组功能。

公开(公告)号：US20160196198A1

公开(公告)日：2016-07-07

申请号：US14590049

申请日：2015-01-06

Applicant: Microsoft Technology Licensing, LLC

Inventor： Harikrishna Menon Ajith Kumar ,  Pankaj Kachrulal Sarda ,  Carlos Pessoa ,  David William Shoots

IPC: G06F11/34 ,  G06F11/32 ,  G06F11/30

CPC classification number: G06F11/3495 ,  G06F11/3003 ,  G06F11/3082 ,  G06F11/323 ,  G06F11/328 ,  G06F11/3476 ,  G06F2201/86 ,  G06F2201/865

Abstract: A performance state machine is controlled in part by identifying notifications from an execution trace of an application program, through rapid automatic comparison of trace events to notification events for notification categories. Some notification categories include application startup, page outline load, page data load start, page data load finish, page to page transition, application input, window size change, media query, binding update, page background task start, page background task finish, developer-defined scenario start, and developer-defined scenario finish. Notifications may reflect heuristics such as the time from startup to first frame submission. A state is placed in the performance state machine for each identified notification, with aggregate application performance data for each transition between identified notifications. Some performance data categories include network activity, disk activity, memory usage, parse time, frame time, dropped frames, component or overall frame rates, and thread utilization. Timelines and other visual representations aid application performance optimization.

Abstract translation: 通过从跟踪事件的快速自动比较到通知类别的通知事件的快速自动比较，通过识别应用程序的执行跟踪的通知来部分地控制性能状态机。 一些通知类别包括应用程序启动，页面大小加载，页面数据加载启动，页面数据加载完成，页面转换，应用程序输入，窗口大小更改，媒体查询，绑定更新，页面后台任务启动，页面后台任务完成，开发人员 定义的场景开始和开发人员定义的场景完成。 通知可能会反映启发式到首次提交时间的启发式。 对于每个识别的通知，状态被放置在性能状态机中，具有用于标识的通知之间的每个转换的聚合应用性能数据。 某些性能数据类别包括网络活动，磁盘活动，内存使用情况，解析时间，帧时间，丢帧，组件或总帧速率以及线程利用率。 时间线和其他视觉表示可以帮助应用程序性能优化。

公开(公告)号：US20160188434A1

公开(公告)日：2016-06-30

申请号：US15066526

申请日：2016-03-10

Applicant: Huawei Technologies Co., Ltd.

Inventor： Huimin Cui ,  Jiacheng Zhao ,  Xiaobing Feng

IPC: G06F11/34

CPC classification number: G06F11/3457 ,  G06F11/3433 ,  G06F11/3452 ,  G06F2201/865

Abstract: A method and a device for determining a program performance interference model is described. The method includes: selecting programs from a determined sample program set to form multiple subsets; acquiring a value of performance interference imposed on each program in each subset and a total occupancy rate of a shared resource occupied by all the programs in each subset; dividing all the subsets into multiple analytical units; performing a regression analysis on the value of performance interference on each sample program included in each analytical unit and a total occupancy rate corresponding to a subset in which the sample program is loaded, and acquiring a target function model; acquiring a performance interference model corresponding to a target program according to the target function model. The performance interference model may be used for preventing another program whose mutual interference is relatively strong from running together with the target program.

Abstract translation: 描述了用于确定节目性能干扰模型的方法和装置。 该方法包括：从确定的样本程序集中选择程序以形成多个子集; 获取施加在每个子集中的每个节目的性能干扰的值以及每个子集中所有节目占用的共享资源的总占用率; 将所有子集划分为多个分析单元; 对每个分析单元中包括的每个样本程序的性能干扰值进行回归分析，以及对应于样本程序加载的子集的总占用率，并获取目标函数模型; 根据目标函数模型获取与目标程序相对应的性能干扰模型。 性能干扰模型可以用于防止相互干扰相对较强的另一个程序与目标程序一起运行。

公开(公告)号：US09379903B1

公开(公告)日：2016-06-28

申请号：US13838433

申请日：2013-03-15

Applicant: Google Inc.

Inventor： Jon P McAlister ,  Justin Christopher Haugh

IPC: G06F15/173 ,  H04L12/24

CPC classification number: H04L41/042 ,  G06F11/302 ,  G06F11/3409 ,  G06F11/3433 ,  G06F2201/81 ,  G06F2201/865

Abstract: A method, system, and computer readable medium for managing applications on an application execution system are disclosed. On an application server the number of instances of a first application type that are in a busy state is determined. This determination is performed at each respective time interval in a plurality of time intervals. Then, a first running average for the busy state of the first application type is computed based upon the number of instances of the first application type that are in a busy state, at the application server, at each respective time interval. A removal request is sent when the first running average for the busy state meets a first removal criterion. The removal request is a request to remove the application server from a data structure that specifies which of a plurality of application servers accept service requests for the first application type.

Abstract translation: 公开了一种用于在应用执行系统上管理应用的方法，系统和计算机可读介质。 在应用服务器上，确定处于忙状态的第一应用类型的实例数。 在多个时间间隔中的每个相应时间间隔执行该确定。 然后，基于处于忙状态的第一应用类型的实例的数量，在应用服务器处，以各自的时间间隔计算第一应用类型的忙状态的第一运行平均值。 当繁忙状态的第一个运行平均值满足第一个删除标准时，将发送删除请求。 删除请求是从数据结构去除应用服务器的请求，该数据结构指定多个应用服务器中的哪一个接受针对第一应用类型的服务请求。

公开(公告)号：US09378472B2

公开(公告)日：2016-06-28

申请号：US12341180

申请日：2008-12-22

Applicant: Rani Kumar ,  Ramanarayanan Krishnaiyer

Inventor： Rani Kumar ,  Ramanarayanan Krishnaiyer

IPC: G06F15/16 ,  G06F9/44 ,  G06Q10/06 ,  G06F11/30 ,  G06F11/34 ,  G06Q30/02

CPC classification number: G06Q10/06 ,  G06F11/3003 ,  G06F11/3072 ,  G06F11/3438 ,  G06F11/3466 ,  G06F2201/86 ,  G06F2201/865 ,  G06Q30/02

Abstract: Methods and systems for tracking user interactions with a computer application. As a computer application is used, it keeps track of user interactions, for example, for use on an analytics server. An interaction tracking configuration may specify which events are tracked, under what conditions the events are tracked, and identify what information is recorded. This configuration may be separated from the application. For example, the configuration may be stored in a configuration file at a location specified within and used by an application. The configuration may then be changed without changing a deployed application. Certain embodiments provide a tracking configuration tool to facilitate the creation of such a configuration. The tool may use a running application to identify events for tracking. Identifying events in this way can simplify the task of configuring interaction tracking by reducing the need for understanding or accessing of the actual code of the application.

Abstract translation: 跟踪用户与计算机应用程序交互的方法和系统。 当使用计算机应用程序时，它会跟踪用户交互，例如用于分析服务器。 交互跟踪配置可以指定哪些事件被跟踪，在什么情况下跟踪事件，并且识别记录什么信息。 该配置可以与应用程序分离。 例如，配置可以存储在配置文件中在应用程序中指定并由应用程序使用的位置。 然后可以改变配置而不改变部署的应用。 某些实施例提供了跟踪配置工具以便于创建这样的配置。 该工具可以使用正在运行的应用程序来识别用于跟踪的事件。 以这种方式识别事件可以通过减少理解或访问应用程序的实际代码的需要来简化配置交互跟踪的任务。

公开(公告)号：US20160179654A1

公开(公告)日：2016-06-23

申请号：US14975902

申请日：2015-12-21

Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION

Inventor： Benjamin D. Cox ,  Daniel P. Craggs ,  Christopher J. Poole

IPC: G06F11/36

CPC classification number: G06F11/3419 ,  G06F11/3466 ,  G06F2201/865

Abstract: Method and system are provided for providing elapsed time indications for source code in a development environment. The method includes: defining blocks of source code to be timed during source code execution; monitoring defined blocks of source code during execution to determine an elapsed time for the execution of the defined block of source code; recording the elapsed time for a defined block of source code; and providing an elapsed time indication for the defined block of source code.

Abstract translation: 提供了方法和系统，用于在开发环境中提供源代码的经过时间指示。 该方法包括：在源代码执行期间定义要定时的源代码块; 在执行期间监视定义的源代码块以确定用于执行所定义的源代码块的经过时间; 记录定义的源代码块的经过时间; 以及为所定义的源代码块提供经过时间指示。

公开(公告)号：US09372773B2

公开(公告)日：2016-06-21

申请号：US13916417

申请日：2013-06-12

Applicant: Advanced Micro Devices, Inc.

Inventor： Joseph Lee Greathouse ,  Anton Chernoff

IPC: G06F9/44 ,  G06F11/30 ,  G06F9/30

CPC classification number: G06F11/30 ,  G06F9/3005 ,  G06F11/3003 ,  G06F11/3471 ,  G06F11/3476 ,  G06F11/3648 ,  G06F2201/865

Abstract: A processor, a method and a computer-readable medium for recording branch addresses are provided. The processor comprises hardware registers and first and second circuitry. The first circuitry is configured to store a first address associated with a branch instruction in the hardware registers. The first circuitry is further configured to store a second address that indicates where the processor execution is redirected to as a result of the branch instruction in the hardware registers. The second circuitry is configured to, in response to a second instruction, retrieve a value of at least one of the registers. The second instruction can be a user-level instruction.

Abstract translation: 提供了用于记录分支地址的处理器，方法和计算机可读介质。 处理器包括硬件寄存器和第一和第二电路。 第一电路被配置为在硬件寄存器中存储与分支指令相关联的第一地址。 第一电路还被配置为作为硬件寄存器中的分支指令的结果存储指示处理器执行被重定向到哪里的第二地址。 第二电路被配置为响应于第二指令检索至少一个寄存器的值。 第二条指令可以是用户级指令。

公开(公告)号：US09372754B2

公开(公告)日：2016-06-21

申请号：US14694213

申请日：2015-04-23

Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION

Inventor： Adam L. Soderlund ,  Wen Wei Tang ,  Song Qing Wu

IPC: G06F11/00 ,  G06F11/14 ,  G06F9/44

CPC classification number: G06F11/1417 ,  G06F9/4406 ,  G06F11/1438 ,  G06F11/1471 ,  G06F2201/865

Abstract: Restoring from a legacy OS environment to a Unified Extensible Firmware Interface (UEFI) pre-boot environment, including: storing, under the UEFI pre-boot environment, context in the UEFI pre-boot environment that needs to be preserved, where the context in the UEFI pre-boot environment that needs to be preserved includes CPU execution context; restoring a first portion of the CPU execution context in response to the UEFI pre-boot environment failing to load the legacy OS; making a CPU associated with the UEFI pre-boot environment enter into System Management Mode, and restoring a second portion of the CPU execution context under the System Management Mode; and exiting from CPU System Management Mode, thereby returning to the UEFI pre-boot environment.

公开(公告)号：US20160170438A1

公开(公告)日：2016-06-16

申请号：US14565512

申请日：2014-12-10

Applicant: Intel Corporation

Inventor： Keqiang WU ,  Jiwei Lu ,  Yong-Fong Lee

IPC: G06F1/12 ,  G06F9/30

CPC classification number: G06F9/30065 ,  G06F8/52 ,  G06F8/70 ,  G06F9/30087 ,  G06F11/3024 ,  G06F11/3409 ,  G06F11/3466 ,  G06F2201/81 ,  G06F2201/825 ,  G06F2201/865 ,  G06F2201/88

Abstract: One embodiment provides an apparatus. The apparatus includes a processor, a chipset, a memory to store a process, and logic. The processor includes one or more core(s) and is to execute the process. The logic is to acquire performance monitoring data in response to a platform processor utilization parameter (PUP) greater than a detection utilization threshold (UT), identify a spin loop based, at least in part, on at least one of a detected hot function and/or a detected hot loop, modify the identified spin loop using binary translation to create a modified process portion, and implement redirection from the identified spin loop to the modified process portion.

Abstract translation: 一个实施例提供了一种装置。 该装置包括处理器，芯片组，用于存储处理的存储器和逻辑。 处理器包括一个或多个核心，并且是执行该过程。 逻辑是响应于大于检测利用阈值（UT）的平台处理器利用参数（PUP）来获取性能监视数据，至少部分地基于检测到的热功能和 /或检测到的热循环，使用二进制转换修改所识别的自旋循环，以创建经修改的处理部分，并且实现从所识别的旋转循环到修改的处理部分的重定向。

公开(公告)号：US09367601B2

公开(公告)日：2016-06-14

申请号：US13843347

申请日：2013-03-15

Applicant: Duke University

Inventor： Shivnath Babu ,  Herodotos Herodotou

IPC: G06F17/30 ,  G06F7/00 ,  G06F11/34 ,  G06F9/50

CPC classification number: G06F17/30595 ,  G06F9/5066 ,  G06F11/3404 ,  G06F11/3409 ,  G06F11/3419 ,  G06F11/3428 ,  G06F11/3442 ,  G06F11/3452 ,  G06F11/3457 ,  G06F11/3466 ,  G06F2201/865

Abstract: Cost-based optimization of configuration parameters and cluster sizing for distributed data processing systems are disclosed. According to an aspect, a method includes receiving at least one job profile of a MapReduce job. The method also includes using the at least one job profile to predict execution of the MapReduce job within a plurality of different predetermined settings of a distributed data processing system. Further, the method includes determining one of the predetermined settings that optimizes performance of the MapReduce job. The method may also include automatically adjusting the distributed data processing system to the determined predetermined setting.

Abstract translation: 公开了用于分布式数据处理系统的配置参数和集群大小的基于成本的优化。 根据一方面，一种方法包括接收MapReduce作业的至少一个作业简档。 该方法还包括使用至少一个作业简档来预测分布式数据处理系统的多个不同预定设置内的MapReduce作业的执行。 此外，该方法包括确定优化MapReduce作业的性能的预定设置之一。 该方法还可以包括将分布式数据处理系统自动调整到所确定的预定设置。