US 11,966,273 B2
Throughput-optimized, quality-of-service aware power capping system
Vasileios Kontorinis, Mountain View, CA (US); Shaohong Li, Santa Clara, CA (US); Xiao Zhang, San Jose, CA (US); Sreekumar Vadakke Kodakara, Campbell, CA (US); and Kunqi Ye, Mountain View, CA (US)
Assigned to Google LLC, Mountain View, CA (US)
Filed by Google LLC, Mountain View, CA (US)
Filed on Feb. 23, 2023, as Appl. No. 18/173,293.
Application 18/173,293 is a continuation of application No. 17/243,853, filed on Apr. 29, 2021, granted, now 11,599,184, issued on Mar. 7, 2023.
Claims priority of provisional application 63/030,639, filed on May 27, 2020.
Prior Publication US 2023/0305618 A1, Sep. 28, 2023
Int. Cl. G06F 1/329 (2019.01); G06F 9/48 (2006.01); G06F 9/50 (2006.01)
CPC G06F 1/329 (2013.01) [G06F 9/4893 (2013.01); G06F 9/5094 (2013.01)] 16 Claims
OG exemplary drawing
 
1. A method, comprising:
receiving, by one or more processors, power measurements for a plurality of machines in a datacenter, the plurality of machines executing one or more tasks;
identifying power limitations for the plurality of machines;
detecting, by the one or more processors, a triggering event for shedding power consumed by the plurality of machines;
commanding, by the one or more processors when the triggering event is detected, the plurality of machines to operate in accordance with one or more restrictions to reduce power consumption;
proactively capping power to the plurality of machines during a period of time following the detected event, the proactive capping based on one or more predictions related to power or load, wherein the triggering event comprises detecting that power measurements are no longer available and wherein the predictions comprise predicted power consumption based on previously received power measurements.