公开(公告)号：US09229766B2

公开(公告)日：2016-01-05

申请号：US14551235

申请日：2014-11-24

Applicant: CA, Inc.

Inventor： Daniel H. Acheff ,  John J. Michelsen ,  James Stephen Kress ,  Steven D. LaRocca

IPC: G06F13/00 ,  G06F9/46 ,  G06F11/36 ,  G06F9/54

CPC classification number: G06F9/466 ,  G06F9/54 ,  G06F11/3006 ,  G06F11/3447 ,  G06F11/3457 ,  G06F11/36 ,  G06F11/3696 ,  G06F2201/815 ,  G06F2201/865 ,  G06F2201/87

Abstract: Request data is identified that describes a request of a second program by a first program intercepted by a first exit in the first program. Response data is identified that describes a response by the second program to the request as intercepted by a second exit in the first program. A virtual service configured to model operation of the second program is instantiated based on the request data and response data. In another aspect, a request by the first program to the second program is intercepted by the first exit and the request is redirected to the virtual service. A response is received as generated by the virtual service and the response is returned to the first program using the second exit.

Abstract translation: 识别通过第一程序中的第一个出口截取的第一个程序来描述第二个程序的请求的请求数据。 识别响应数据，其描述第二程序对第一程序中的第二个出口截取的请求的响应。 基于请求数据和响应数据来实例化配置成对第二程序进行建模的虚拟服务。 在另一方面，第一程序对第二程序的请求被第一个退出截取，并且请求被重定向到虚拟服务。 接收到由虚拟服务产生的响应，并且使用第二个退出将响应返回到第一个程序。

公开(公告)号：US20150378872A1

公开(公告)日：2015-12-31

申请号：US14316191

申请日：2014-06-26

Applicant: GOOGLE INC.

Inventor： Keun Soo KIM

IPC: G06F11/36

CPC classification number: G06F11/3664 ,  G06F11/3466 ,  G06F11/3688 ,  G06F2201/865

Abstract: A method for automated parallel testing of a software application using value replication is disclosed. An example embodiment may include marking at least one event in the software application with a primary marker that includes a primary marker label and a primary marker configuration value. The primary marker may be replicated to create a replicate marker that has a replicate marker label and a replicate configuration value. The replicate marker label and configuration value may be different from the primary marker label and configuration value. Parallel executions of the software application may be performed using the primary and replicate markers so that the software application is executed with both the primary and replicate marker configuration values. The disclosed middleware-level method transparently implements such replication of the output marker values. That is to make any MapReduce-style applications easily take advantage of output value replication.

Abstract translation: 公开了一种使用价值复制的软件应用程序的自动并行测试方法。 示例实施例可以包括使用包括主标记标签和主标记配置值的主标记来标记软件应用中的至少一个事件。 可以复制主标记以创建具有复制标记标签和复制配置值的复制标记。 复制标记标签和配置值可能与主标记标签和配置值不同。 可以使用主要和复制标记来执行软件应用程序的并行执行，以便使用主要和复制标记配置值来执行软件应用程序。 所公开的中间件级方法透明地实现了输出标记值的这种复制。 那就是让MapReduce风格的应用程序轻松利用输出值复制。

公开(公告)号：US09223571B2

公开(公告)日：2015-12-29

申请号：US14614226

申请日：2015-02-04

Applicant: Infosys Limited

Inventor： Sayantan Mitra ,  Santonu Sarkar

IPC: G06F9/44 ,  G06F9/54 ,  G06F9/45 ,  G06F11/34

CPC classification number: G06F8/77 ,  G06F8/443 ,  G06F8/71 ,  G06F9/54 ,  G06F11/34 ,  G06F11/3409 ,  G06F11/3419 ,  G06F11/3447 ,  G06F11/3452 ,  G06F11/3466 ,  G06F2201/865

Abstract: The present disclosure includes, in a heterogeneous system, receiving a desired speedup of an application as input and performing a static analysis and a dynamic analysis of the application. The dynamic analysis of the application comprises, identifying a set of parameters including, an end-to-end execution time of the application, an execution time of data parallel loops in the application, an execution time of non-data parallel loops in the application, and an amount of physical memory used by each data structure in each data parallel loop. Dynamic analysis also includes calculating and providing the feasibility of achieving the desired speedup of the application based on the identified set of parameters, and satisfaction of each of, an initialization invariant, a data-parallel invariant and a data transfer invariant.

Abstract translation: 本公开包括在异构系统中接收作为输入的应用的期望加速度并且执行应用的静态分析和动态分析。 应用程序的动态分析包括：识别一组参数，包括应用程序的端到端执行时间，应用程序中数据并行循环的执行时间，应用程序中非数据并行循环的执行时间 ，以及每个数据并行循环中每个数据结构使用的物理内存量。 动态分析还包括基于所识别的参数集合来计算和提供实现所需应用加速度的可行性，以及初始化不变量，数据并行不变量和数据传输不变量的满足度。

公开(公告)号：US20150355940A1

公开(公告)日：2015-12-10

申请号：US14828795

申请日：2015-08-18

Applicant: International Business Machines Corporation

Inventor： Jonathan D. Bradbury ,  Fadi Y. Busaba ,  Mark S. Farrell ,  Charles W. Gainey, JR. ,  Dan F. Greiner ,  Lisa Cranton Heller ,  Jeffrey P. Kubala ,  Damian L. Osisek ,  Donald W. Schmidt ,  Timothy J. Slegel

IPC: G06F9/48 ,  G06F9/46

CPC classification number: G06F11/3423 ,  G06F1/14 ,  G06F9/30076 ,  G06F9/30145 ,  G06F9/3851 ,  G06F9/45533 ,  G06F9/46 ,  G06F9/4887 ,  G06F9/5077 ,  G06F11/302 ,  G06F11/3024 ,  G06F11/3442 ,  G06F2201/865

Abstract: Embodiments relate to idle time accumulation in a multithreading computer system. According to one aspect, a computer-implemented method for idle time accumulation in a computer system is provided. The computer system includes a configuration having a plurality of cores and an operating system (OS)-image configurable between single thread (ST) mode and a multithreading (MT) mode in a logical partition. The MT mode supports multiple threads on shared resources per core simultaneously. The method includes executing a query instruction on an initiating core of the plurality of cores. The executing includes obtaining, by the OS-image, a maximum thread identification value indicating a current maximum thread identifier of the cores within the logical partition. The initiating core also obtains a multithreading idle time value for each of the cores indicating an aggregate amount of idle time of all threads enabled on each of the cores in the MT mode.

Abstract translation: 实施例涉及多线程计算机系统中的空闲时间累积。 根据一个方面，提供了一种用于计算机系统中的空闲时间累积的计算机实现的方法。 计算机系统包括具有多个核心的配置和在逻辑分区中在单线程（ST）模式和多线程（MT））模式之间可配置的操作系统（OS）图像。 MT模式同时支持每个核心共享资源上的多个线程。 该方法包括在多个核的启动核上执行查询指令。 执行包括通过OS映像获得指示逻辑分区内的核的当前最大线程标识符的最大线程标识值。 启动核心还获得每个核心的多线程空闲时间值，其指示在MT模式中在每个核心上启用的所有线程的空闲时间的总量。

公开(公告)号：US09208053B2

公开(公告)日：2015-12-08

申请号：US14334034

申请日：2014-07-17

Applicant: Yahoo! Inc.

Inventor： Arun Kejariwal ,  Paolo D'Alberto ,  Sapan Panigrahi ,  Rosario Cammarota

IPC: G06F9/44 ,  G06F9/45 ,  G06F11/34

CPC classification number: G06F11/3447 ,  G06F11/3409 ,  G06F11/3419 ,  G06F11/3442 ,  G06F11/3466 ,  G06F2201/81 ,  G06F2201/83 ,  G06F2201/86 ,  G06F2201/865 ,  G06F2201/88

Abstract: A system and method for identifying optimal system architectures for a reference application are provided. The system and method comprise executing a reference application and a plurality of test applications on a current system architecture and sampling performance data for each of the applications. The performance data is used to compute an application signature for each application. A similarity element is derived from the application signatures that illustrates the similarity between each application and every other application. Using a similarity threshold and an algorithm, a subset of test applications that are similar to the reference application are derived.

Abstract translation: 提供了一种用于识别参考应用的最佳系统架构的系统和方法。 该系统和方法包括在当前系统架构上执行参考应用和多个测试应用，并针对每个应用采样性能数据。 性能数据用于计算每个应用程序的应用程序签名。 相似性元素从应用程序签名中得出，该签名说明了每个应用程序和每个其他应用程序之间的相似性。 使用相似性阈值和算法，推导出与参考应用程序相似的测试应用程序子集。

公开(公告)号：US20150347282A1

公开(公告)日：2015-12-03

申请号：US14577806

申请日：2014-12-19

Applicant: Apple Inc.

Inventor： Joar Wingfors ,  Andrew J. Nesbitt ,  Brent Robert Shank ,  Brooke KaleoOKalani Callahan ,  Kevin Charles Milden ,  Maxwell Oliver Drukman

IPC: G06F11/36 ,  G06F11/34

CPC classification number: G06F11/3688 ,  G06F11/3409 ,  G06F11/3452 ,  G06F11/3664 ,  G06F11/3692 ,  G06F2201/865

Abstract: Systems, methods, and computer-readable storage media for testing performance of blocks of code. A system can first receive a request to run a performance test for a selected portion of code in an application. Next, the system can obtain a baseline performance test result corresponding to the performance test for the selected portion of code in the application, and run the performance test for the selected portion of code in the application. The system can then generate a performance test result based on the performance test and the baseline performance test result. The system can also present the test result and/or the baseline performance test result via a graphical user interface.

Abstract translation: 用于测试代码块性能的系统，方法和计算机可读存储介质。 系统可以首先接收对应用程序中所选部分代码运行性能测试的请求。 接下来，系统可以获得与应用程序中所选部分代码的性能测试相对应的基准性能测试结果，并对应用程序中选定的代码部分运行性能测试。 然后，系统可以基于性能测试和基准性能测试结果生成性能测试结果。 该系统还可以通过图形用户界面呈现测试结果和/或基准性能测试结果。

公开(公告)号：US09201770B1

公开(公告)日：2015-12-01

申请号：US14140721

申请日：2013-12-26

Applicant: EMC Corporation

Inventor： Bryan Duerk

IPC: G06F9/44 ,  G06F11/36

CPC classification number: G06F11/3452 ,  G06F11/3003 ,  G06F11/302 ,  G06F11/3438 ,  G06F2201/865

Abstract: In a computer a graphical user interface (GUI) portion of an application program is instantiated as a core module and variable modules. One of a set of distinct versions of a variable module is selected and for interoperating with the core module, where the versions differ by respective implementations of a user-visible feature. Usage of the GUI involving the user-visible feature is monitored, recorded and reported, including collection of performance information for operations potentially affected by the use of the selected version rather than another version of the variable module. The performance information is transmitted to an evaluation system that collects performance information from a sample population of systems using all the distinct versions of the variable module, and applies statistical analysis to the performance information to yield statistical performance measures for the different implementations, usable to decide whether wider-scale official deployment of a new implementation is desirable.

Abstract translation: 在计算机中，应用程序的图形用户界面（GUI）部分被实例化为核心模块和可变模块。 选择一组不同版本的可变模块之一，并且与核心模块进行互操作，其中版本由用户可见功能的各个实现而不同。 监视，记录和报告涉及用户可见功能的GUI的使用，包括可能受使用所选版本而不是另一版本的可变模块影响的操作的性能信息的收集。 性能信息被传输到评估系统，该系统从使用所有可变模块的所有不同版本的系统的采样群收集性能信息，并将统计分析应用于性能信息以产生用于不同实现的统计性能测量，可用于决定 是否需要更广泛地官方部署新的实施。

公开(公告)号：US09189363B2

公开(公告)日：2015-11-17

申请号：US12900393

申请日：2010-10-07

Applicant: Gregory William Dalcher

Inventor： Gregory William Dalcher

IPC: H04L29/06 ,  G06F21/00 ,  G06F11/34

CPC classification number: G06F21/577 ,  G06F11/3037 ,  G06F11/3409 ,  G06F11/3466 ,  G06F2201/865

Abstract: A system, method, and computer program product are provided for monitoring an execution flow of a function. In use, data associated with a function is identified within a call stack. Additionally, a call stack frame is determined from freed memory in the call stack. Further, an execution flow of the function is monitored, utilizing the call stack frame from the freed memory.

Abstract translation: 提供了用于监视功能的执行流程的系统，方法和计算机程序产品。 在使用中，与调用堆栈中的功能相关联的数据被识别。 另外，从调用堆栈中的释放的存储器确定一个调用堆栈帧。 此外，利用来自释放的存储器的调用堆栈帧来监视该功能的执行流程。

公开(公告)号：US20150317222A1

公开(公告)日：2015-11-05

申请号：US14268046

申请日：2014-05-02

Applicant: International Business Machines Corporation

Inventor： Ruchi Mahindru ,  Harigovind V. Ramasamy ,  Soumitra Sarkar

IPC: G06F11/20 ,  H04L29/08

CPC classification number: G06F11/2033 ,  G06F11/2023 ,  G06F11/2025 ,  G06F11/2028 ,  G06F11/2048 ,  G06F11/2097 ,  G06F11/3003 ,  G06F11/302 ,  G06F2201/805 ,  G06F2201/815 ,  G06F2201/865 ,  H04L67/10 ,  H04L67/1095 ,  H04L67/1097 ,  H04L69/40

Abstract: During normal operation, at a first site, of a disaster recovery management unit, at least one customer workload machine, at least one management service machine, and metadata for the at least one management service machine are replicated to a remote disaster recovery site. After a disaster at the first site, a replicated version of the at least one customer workload machine and a replicated version of the at least one management service machine are brought up at the remote disaster recovery site. A replicated version of the metadata for the at least one management service machine is reconfigured by executing, on the replicated version of the at least one management service machine, a failover script, to obtain reconfigured replicated metadata for the replicated version of the at least one management service machine. When the first site comes back up, failback is carried out, essentially in the reverse order.

Abstract translation: 在正常操作期间，在灾难恢复管理单元的第一站点，至少一个客户工作负载机器，至少一个管理服务机器以及用于至少一个管理服务机器的元数据被复制到远程灾难恢复站点。 在第一站点发生灾难之后，至少一个客户工作负载计算机的复制版本和至少一个管理服务机器的复制版本在远程灾难恢复站点被提出。 通过在所述至少一个管理服务机器的复制版本上执行故障切换脚本来重新配置用于所述至少一个管理服务机器的元数据的复制版本，以获得用于所述至少一个管理服务机器的复制版本的重新配置的复制元数据 管理服务机。 当第一个站点重新启动时，执行故障恢复，基本上是相反的顺序。

公开(公告)号：US09176841B2

公开(公告)日：2015-11-03

申请号：US13339358

申请日：2011-12-28

Applicant: Aman Kansal ,  Ranveer Chandra ,  Jie Liu ,  Paramvir Bahl

Inventor： Aman Kansal ,  Ranveer Chandra ,  Jie Liu ,  Paramvir Bahl

IPC: G06F9/44 ,  G06F11/34 ,  G06F11/30

CPC classification number: G06F11/3457 ,  G06F11/3062 ,  G06F2201/865 ,  Y02D10/34

Abstract: Embodiments provide implementations for estimating a test application's energy usage on a target device based on execution of the test application. Resource usage associated with the execution of the test application is measured. The measured resource usage is input into a power model of the target device to estimate the energy that the test application uses when executed on the target device. An emulation system is configured to execute the test application in a virtual execution environment using resource scaling and simulated target device application programming interfaces to substantially simulate the corresponding capabilities of the target device.

Abstract translation: 实施例提供了基于测试应用的执行来估计测试应用在目标设备上的能量使用的实现。 测量与执行测试应用程序相关联的资源使用情况。 测量的资源使用量被输入到目标设备的功率模型中，以估计在目标设备上执行时测试应用使用的能量。 仿真系统被配置为使用资源缩放和模拟的目标设备应用编程接口在虚拟执行环境中执行测试应用，以基本上模拟目标设备的对应能力。