Distributing power to heterogeneous compute elements of a processor
    73.
    发明授权
    Distributing power to heterogeneous compute elements of a processor 有权
    将功率分配给处理器的异构计算元素

    公开(公告)号:US09335804B2

    公开(公告)日:2016-05-10

    申请号:US13783986

    申请日:2013-03-04

    申请人: Intel Corporation

    IPC分类号: G06F1/32 G06F1/26

    摘要: In one embodiment, the present invention includes a processor having a first domain with a first compute engine and a second domain with a second compute engine, where each of these domains can operate at an independent voltage and frequency. A first logic may be present to update a power bias value used to control dynamic allocation of power between the first and second domains based at least in part on a busyness of the second domain. In turn, a second logic may dynamically allocate at least a portion of a power budget for the processor between the domains based at least in part on this power bias value. Other embodiments are described and claimed.

    摘要翻译: 在一个实施例中,本发明包括处理器,其具有具有第一计算引擎的第一域和具有第二计算引擎的第二域,其中这些域中的每一个可以以独立的电压和频率工作。 可以存在第一逻辑来至少部分地基于第二域的繁忙来更新用于控制第一和第二域之间的功率的动态分配的功率偏置值。 反过来,第二逻辑可以至少部分地基于该功率偏差值来在区域之间动态分配处理器的功率预算的至少一部分。 描述和要求保护其他实施例。

    Method, Apparatus, And System For Energy Efficiency And Energy Conservation Including Power And Performance Balancing Between Multiple Processing Elements And/Or A Communication Bus
    74.
    发明申请
    Method, Apparatus, And System For Energy Efficiency And Energy Conservation Including Power And Performance Balancing Between Multiple Processing Elements And/Or A Communication Bus 审中-公开
    能量效率和节能的方法,装置和系统,包括多个处理元件之间的功率和性能平衡和/或一个通信总线

    公开(公告)号:US20150046730A1

    公开(公告)日:2015-02-12

    申请号:US14526040

    申请日:2014-10-28

    申请人: Intel Corporation

    IPC分类号: G06F1/32 G06F1/20 G06F9/50

    摘要: An apparatus, method and system is described herein for efficiently balancing performance and power between processing elements based on measured workloads. If a workload of a processing element indicates that it is a bottleneck, then its performance may be increased. However, if a platform or integrated circuit including the processing element is already operating at a power or thermal limit, the increase in performance is counterbalanced by a reduction or cap in another processing elements performance to maintain compliance with the power or thermal limit. As a result, bottlenecks are identified and alleviated by balancing power allocation, even when multiple processing elements are operating at a power or thermal limit.

    摘要翻译: 这里描述了一种装置,方法和系统,用于基于测量的工作负载有效地平衡处理元件之间的性能和功率。 如果处理元件的工作负载表明它是瓶颈,则可能会增加其性能。 然而,如果包括处理元件的平台或集成电路已经在功率或热限制下运行,则性能的提高通过另一处理元件的性能的降低或上限来平衡,以保持与功率或热限制的一致性。 因此,即使在多个处理元件在功率或热限制下工作时,通过平衡功率分配来识别和缓解瓶颈。