知搜-全球专利检索

  1. 登录
  2. 注册

搜索结果

960 条记录 (耗时 0.035 秒)

    System, method, and apparatus for improving application-launch latencies
    71.
    发明授权
    System, method, and apparatus for improving application-launch latencies 有权
    用于改进应用启动延迟的系统，方法和设备

    公开(公告)号：US09250958B2

    公开(公告)日：2016-02-02

    申请号：US13681201

    申请日：2012-11-19

    申请人： Qualcomm Innovation Center, Inc.

    发明人： John L. Mark ,  Michael U. Schwartz

    IPC分类号： G06F9/45 ,  G06F9/48 ,  G06F9/50 ,  H04W4/02

    CPC分类号： G06F9/485 ,  G06F9/5022 ,  H04W4/027

    摘要： Methods and systems for launching applications on a computing device are disclosed. An exemplary method may include gathering data in connection with a launch of each of the applications that indicates one or more events occurring in connection with the launch of each application. When an indication of a particular event occurring is received, one or more out-of-memory values are modified based upon the data and the particular event. The out-of-memory values are indicative of a priority in which corresponding background applications are killed from running in the background of the computing device when memory runs low. When memory runs low on the computing device, one or more of the background applications are killed based upon the out-of-memory values to leave background applications running that a user is more likely to launch than the killed applications.

    摘要翻译： 公开了在计算设备上启动应用的方法和系统。 一个示例性的方法可以包括结合指示与每个应用的启动相关联地发生的一个或多个事件的每个应用的启动而收集数据。 当接收到特定事件发生的指示时，基于数据和特定事件来修改一个或多个超出存储器值的值。 内存不足值表示当存储器运行较低时相应的后台应用程序在计算设备的后台运行时被杀死的优先级。 当计算设备上的内存不足时，一个或多个后台应用程序将基于内存不足值进行死机，以使运行用户比发生的应用程序更有可能启动的后台应用程序。

    COMPUTER SYSTEM AND MEMORY ALLOCATION ADJUSTMENT METHOD FOR COMPUTER SYSTEM
    72.
    发明申请
    COMPUTER SYSTEM AND MEMORY ALLOCATION ADJUSTMENT METHOD FOR COMPUTER SYSTEM 有权
    计算机系统的计算机系统和存储器分配调整方法

    公开(公告)号：US20150378892A1

    公开(公告)日：2015-12-31

    申请号：US14766608

    申请日：2013-09-26

    申请人： HITACHI, LTD.

    发明人： Sachie TAJIMA ,  Tadashi TAKEUCHI

    IPC分类号： G06F12/08 ,  G06F9/455

    CPC分类号： G06F12/0802 ,  G06F9/45558 ,  G06F9/5022 ,  G06F12/0848 ,  G06F2009/45583 ,  G06F2212/152 ,  G06F2212/6042

    摘要： In a computer system in which a virtualization control unit controls a plurality of virtual machines, if memory is collected regardless of a memory usage status of the virtual machine, cache miss increases and an IO performance of overall system deteriorates. In order to solve this problem, a usage status of a cache region within a memory which is utilized by each OS that the plurality of the virtual machines has, and based on a monitoring result, the virtualization control unit decides an allocation region of the memory as a collection target among the allocation region of the memory already allocated to each OS, and collects the allocation region of the memory as the collection target from the OS as a current allocation destination.

    摘要翻译： 在其中虚拟化控制单元控制多个虚拟机的计算机系统中，如果存储器被收集而不管虚拟机的存储器使用状态如何，则高速缓存未命中增加，并且整个系统的IO性能恶化。 为了解决这个问题，由多个虚拟机具有的每个OS使用的存储器内的高速缓存区域的使用状态，并且基于监视结果，虚拟化控制单元决定存储器的分配区域 作为已经分配给每个OS的存储器的分配区域中的收集目标，并且从作为当前分配目的地的OS收集作为收集目标的存储器的分配区域。

    Systems and methods for adaptive integration of hardware and software lock elision techniques
    75.
    发明授权
    Systems and methods for adaptive integration of hardware and software lock elision techniques 有权
    硬件和软件锁定技术自适应集成的系统和方法

    公开(公告)号：US09183043B2

    公开(公告)日：2015-11-10

    申请号：US14254758

    申请日：2014-04-16

    申请人： Oracle International Corporation

    发明人： David Dice ,  Alex Kogan ,  Yosef Lev ,  Timothy M. Merrifield ,  Mark S. Moir

    IPC分类号： G06F9/46 ,  G06F9/48 ,  G06F9/50 ,  G06F9/52

    CPC分类号： G06F9/4843 ,  G06F9/467 ,  G06F9/468 ,  G06F9/4881 ,  G06F9/5022 ,  G06F9/528

    摘要： Particular techniques for improving the scalability of concurrent programs (e.g., lock-based applications) may be effective in some environments and for some workloads, but not others. The systems described herein may automatically choose appropriate ones of these techniques to apply when executing lock-based applications at runtime, based on observations of the application in the current environment and with the current workload. In one example, two techniques for improving lock scalability (e.g., transactional lock elision using hardware transactional memory, and optimistic software techniques) may be integrated together. A lightweight runtime library built for this purpose may adapt its approach to managing concurrency by dynamically selecting one or more of these techniques (at different times) during execution of a given application. In this Adaptive Lock Elision approach, the techniques may be selected (based on pluggable policies) at runtime to achieve good performance on different platforms and for different workloads.

    摘要翻译： 用于提高并发程序（例如基于锁的应用程序）的可扩展性的特殊技术在一些环境中以及对于一些工作负载而言可能是有效的，而不是其他工作负载。 基于当前环境中的应用和当前工作负载的观察，本文所述的系统可以自动选择在运行时执行基于锁的应用时应用的这些技术中适当的系统。 在一个示例中，可以集成两种用于提高锁可伸缩性的技术（例如，使用硬件事务存储器的事务锁定检测和乐观软件技术）。 为此目的而构建的轻量级运行时库可以通过在执行给定应用程序期间动态选择这些技术（在不同时间）中的一个或多个技术来调整其方法来管理并发性。 在这种自适应锁定Elision方法中，可以在运行时选择（基于可插拔策略）的技术，以在不同的平台和不同的工作负载下实现良好的性能。

    Memory management techniques
    76.
    发明授权
    Memory management techniques 有权
    内存管理技术

    公开(公告)号：US09158699B2

    公开(公告)日：2015-10-13

    申请号：US14549501

    申请日：2014-11-20

    申请人： Microsoft Corporation

    发明人： Steve Pronovost ,  Maxwell Abernethy ,  Rudolph Balaz ,  Ameet Chitre

    IPC分类号： G06F12/02 ,  G06F12/08 ,  G09G5/00 ,  G06F12/06 ,  G06F9/50 ,  G06T1/60

    CPC分类号： G06F12/023 ,  G06F9/5016 ,  G06F9/5022 ,  G06F12/02 ,  G06F12/0638 ,  G06F12/0891 ,  G06F2212/1044 ,  G06F2212/657 ,  G06T1/60 ,  G09G5/001

    摘要： Memory management techniques that permit an executing process to store content in memory and later retrieve that content from the memory, but that also permit a memory manager to discard that content to address memory pressure. A process executing on a computing device may notify a memory manager of the computing device that first memory space allocated to the process contains first content that is available for discard. If the memory manager detects the computing device is experiencing memory pressure, the memory manager may address the memory pressure by selecting memory space available for discard and discarding the content of the memory space. Before a process reuses content made available for discard, the process may notify the memory manager of the intent to reuse and, in response, receive empty memory and an indication that the content was discarded or receive an indication that the content is still available for use.

    摘要翻译： 存储器管理技术允许执行过程将内容存储在存储器中，并且随后从存储器检索该内容，但也允许存储器管理器丢弃该内容以解决存储器压力。 在计算设备上执行的过程可以向计算设备的存储管理器通知分配给该过程的第一存储器空间包含可用于丢弃的第一内容。 如果存储器管理器检测到计算设备正在经历存储器压力，则存储器管理器可以通过选择可用于丢弃并丢弃存储器空间的内容的存储器空间来解决存储器压力。 在过程重新使用可用于丢弃的内容之前，该过程可以通知存储器管理器重新使用的意图，并且作为响应，接收空的存储器以及内容被丢弃的指示或者接收到内容仍然可用的指示 。

    Immediately launching applications
    77.
    发明授权
    Immediately launching applications 有权
    立即启动应用程序

    公开(公告)号：US09158577B2

    公开(公告)日：2015-10-13

    申请号：US13569635

    申请日：2012-08-08

    申请人： James Andrew Trenton Lipscomb ,  Christopher L. McGilliard

    发明人： James Andrew Trenton Lipscomb ,  Christopher L. McGilliard

    IPC分类号： G06F15/173 ,  G06F9/48 ,  G06F9/50

    CPC分类号： G06F8/656 ,  G06F9/485 ,  G06F9/5022 ,  G06F9/5083 ,  G06F9/5088 ,  H04L67/34

    摘要： Disclosed are various embodiments for a deployment management system. A second version of a deployable application is executed concurrently with a first version. Network traffic sent to the first version of the application is redirected to the second version. In the event of an error, network traffic is directed back to the first version of the application. After a period of concurrent execution, the first version of the application is terminated.

    摘要翻译： 公开了用于部署管理系统的各种实施例。 可部署应用程序的第二个版本与第一个版本同时执行。 发送到应用程序的第一个版本的网络流量被重定向到第二个版本。 如果发生错误，网络流量将返回到应用程序的第一个版本。 经过一段时间的并发执行，应用程序的第一个版本被终止。

    Opportunistically scheduling and adjusting time slices
    80.
    发明授权
    Opportunistically scheduling and adjusting time slices 有权
    机会调度和调整时间片

    公开(公告)号：US09086922B2

    公开(公告)日：2015-07-21

    申请号：US12605953

    申请日：2009-10-26

    申请人： Thomas Fahrig ,  David Cutler

    发明人： Thomas Fahrig ,  David Cutler

    IPC分类号： G06F9/46 ,  G06F9/50 ,  G06F9/52

    CPC分类号： G06F9/4887 ,  G06F9/5022 ,  G06F9/5038 ,  G06F9/5077 ,  G06F9/526 ,  G06F2209/506

    摘要： Computerized methods, computer systems, and computer-readable media for governing how virtual processors are scheduled to particular logical processors are provided. A scheduler is employed to balance a load imposed by virtual machines, each having a plurality of virtual processors, across various logical processors (comprising a physical machine) that are running threads in parallel. The threads are issued by the virtual processors and often cause spin waits that inefficiently consume capacity of the logical processors that are executing the threads. Upon detecting a spin-wait state of the logical processor(s), the scheduler will opportunistically grant time-slice extensions to virtual processors that are running a critical section of code, thus, mitigating performance loss on the front end. Also, the scheduler will mitigate performance loss on the back end by opportunistically de-scheduling then rescheduling a virtual machine in a spin-wait state to render the logical processor(s) available for other work in the interim.

    摘要翻译： 提供了用于管理如何将虚拟处理器调度到特定逻辑处理器的计算机化方法，计算机系统和计算机可读介质。 采用调度器来平衡在并行运行线程的各种逻辑处理器（包括物理机器）上的虚拟机施加的负载，每个虚拟机具有多个虚拟处理器。 线程由虚拟处理器发出，并且经常引起旋转等待，这些等待无效率地消耗正在执行线程的逻辑处理器的容量。 在检测到逻辑处理器的旋转等待状态时，调度器将机会地向运行临界代码段的虚拟处理器授予时间片扩展，从而减轻前端的性能损失。 此外，调度器将通过机会地去调度来减轻后端的性能损失，然后在旋转等待状态下重新安排虚拟机以使逻辑处理器可用于临时中的其他工作。