A METHOD FOR DETERMINING TRANSITION POINTS ON MULTIPLE PERFORMANCE STATE CAPABLE MICROPROCESSORS

    公开(公告)号:WO2002074046A3

    公开(公告)日:2002-09-26

    申请号:PCT/US2002/002028

    申请日:2002-01-24

    Abstract: A method for automatically transitioning a processor to another performance level in a demang-based system. The invention provides for the automatic adjustment of processor frequency while preserving system responsiveness. The performance-level policy algorithm of the present invention detects increased processor utilization quickly enough that tranistion to a higher performance level is comparable to maximum system performance. The performance-level is comparable to maximum system performance. The performance-level policy algorithm of an embodiment of the present invention delays processor transition to a lower performance level so that quick reversals in demand do not precipitate unnecessary transitioning.

    TECHNIQUES FOR PLATFORM DUTY CYCLING
    3.
    发明申请
    TECHNIQUES FOR PLATFORM DUTY CYCLING 审中-公开
    平台责任周期的技术

    公开(公告)号:WO2014105203A1

    公开(公告)日:2014-07-03

    申请号:PCT/US2013/048427

    申请日:2013-06-28

    Abstract: Various embodiments are generally directed to an apparatus, method and other techniques for detecting active and semi-active workloads during execution on a platform processing device and enabling a duty cycle process to reduce thermal output and power consumption, and align unaligned activity. In various embodiments, the duty cycle processing may be enabled during an active workload when thermal output or power consumption is above a thermal threshold or power consumption threshold that is below an efficient operating point for the platform processing device. The duty cycle processing may also be enabled during semi-active workloads when the workload causes the platform processing device to be underutilized and unaligned. The duty cycle processing may comprise enabling a forced idle period for the platform processing device. Other embodiments are described and claimed.

    Abstract translation: 各种实施例通常涉及用于在平台处理设备上的执行期间检测活动和半主动工作负载的装置,方法和其他技术,并且实现占空比过程以减少热输出和功率消耗,并且调整未对齐的活动。 在各种实施例中,当热输出或功率消耗高于低于用于平台处理设备的有效工作点的热阈值或功率消耗阈值时,可以在活动工作负载期间启用占空比处理。 当工作负载导致平台处理设备未充分利用和未对齐时,也可以在半主动工作负载期间启用占空比处理。 占空比处理可以包括为平台处理装置启用强制空闲周期。 描述和要求保护其他实施例。

    EXTENDED THERMAL MANAGEMENT
    4.
    发明申请
    EXTENDED THERMAL MANAGEMENT 审中-公开
    扩展热管理

    公开(公告)号:WO2006011992A2

    公开(公告)日:2006-02-02

    申请号:PCT/US2005020975

    申请日:2005-06-14

    CPC classification number: G06F1/206 Y02D10/16

    Abstract: A method for managing thermal condition of a thermal zone that includes multiple thermally controllable components include determining thermal relationship between the components and reducing temperature of a first component by reducing thermal dissipation of a second component.

    Abstract translation: 用于管理包括多个可热控制部件的热区的热状态的方法包括通过减少第二部件的散热来确定部件之间的热关系并降低第一部件的温度。

    A METHOD FOR DETERMINING TRANSITION POINTS ON MULTIPLE PERFORMANCE STATE CAPABLE MICROPROCESSORS
    9.
    发明申请
    A METHOD FOR DETERMINING TRANSITION POINTS ON MULTIPLE PERFORMANCE STATE CAPABLE MICROPROCESSORS 审中-公开
    用于确定多个性能状态可编程微处理器的过渡点的方法

    公开(公告)号:WO2002074046A2

    公开(公告)日:2002-09-26

    申请号:PCT/US2002/002028

    申请日:2002-01-24

    Abstract: A method for automatically transitioning a processor to another performance level in a demang-based system. The invention provides for the automatic adjustment of processor frequency while preserving system responsiveness. The performance-level policy algorithm of the present invention detects increased processor utilization quickly enough that tranistion to a higher performance level is comparable to maximum system performance. The performance-level is comparable to maximum system performance. The performance-level policy algorithm of an embodiment of the present invention delays processor transition to a lower performance level so that quick reversals in demand do not precipitate unnecessary transitioning.

    Abstract translation: 一种在基于demang的系统中自动将处理器转换到另一个性能级别的方法。 本发明提供了处理器频率的自动调整,同时保持系统响应性。 本发明的性能级别策略算法足够快地检测提高的处理器利用率,使得到更高性能级别的转换与最大系统性能相当。 性能级别与最大系统性能相当。 本发明的实施例的性能级别策略算法将处理器转换延迟到较低的性能水平,使得需求的快速反转不会引起不必要的转换。

Patent Agency Ranking