International Business Machines Corporation

Todd Jon Rosedahl

Malcolm Scott Allen-Ware

Karthick Rajamani

Guillermo Jesus Silva

Gregory Scott Still

Charles Robert Lefurgy

Timothy Glen Hallett

Jordan Ross Keuseman

Martha Broyles

Patrick Kevin Egan

Heather Lynn Hanson

Victor V Zyuban

Glenn Rueban Miles

John Steven Dodson

Alan J Drake

Srinivasan Ramani

Kirk David Peterson

Michael Stephen Floyd

Gary A Van Huben

Eric Eugene Retter

Scot H Rider

Neil Clair Berglund

James Douglas Jordan

Irving G Baysah

Benjamin W Mashak

Alper Buyuktosunoglu

Brian William Thompto

Tilman Gloekler

Bishop Chapman Brock

Mark Robert Funk

Philip Lee Vitale

Douglas Michael Boecker

Christopher Francois

Joshua D Friedrich

Allan E Johnson

Pradip Bose

Malcolm Scott Ware

Wei T Huang

Freeman Leigh Rawson, Iii

Hans Mikael Jacobson

Kenneth Lee Wright

William James Anderl

John Bruce Carter

Christopher Landon Wood

Augusto Javier Vega

Brent William Jacobs

Howard V Mahaney, Jr

Raymond J Harrington

Michael Lee Miller

Joseph Edward Bolan

Eric Jason Fluhr

Eun Kyung Lee

Patrick Leo Caporale

Christopher Harry Austen

Peter Adam Wendling

Steven William Steele

Eric D Marz

Jeffrey Scot Rotter

Derek Howard

Greg Abrami

Fadi M Kassem

Malcom S Allen-Ware

Thermal control of a multi-chip module in an operating environment is facilitated by predetermining separate thermal control points for multiple chips of the multi-chip module, with a first chip and a second chip having different predetermined thermal control points, and saving the predetermined thermal control points for reference by a thermal control of the multi-chip module in an operating environment. The thermal control monitors an operating temperature of the first chip, and compares the operating temperature of the first chip to the predetermined thermal control point of that chip. The thermal control further initiates a control action to control temperature of the first chip based on comparing the operating temperature of the first chip to the predetermined thermal control point of the first chip.

Predetermining separate thermal control points for chips of a multi-chip module

A power controller that controls power usage of a heterogeneous system is provided. The power controller receives a power usage cap for a heterogeneous system that includes multiple components. The power controller assigns priorities that respectively correspond to the multiple components. The power controller assigns power shifting ratios that respectively correspond to the multiple components. The power controller adjusts a total power usage of the multiple components to be within a threshold of the power usage cap by adjusting the power usage of individual components in a sequence defined by the assigned priorities. The power controller shifts a power usage from a first-priority component to a second-priority component (and so on) according to the assigned power shifting ratios.

Power shifting among hardware components in heterogeneous system

A distributed power management system is configured determine a node power consumption of a node during a first time interval. The system can determine a node power cap. The system can determine a proportional component power budget for a component of the node based, at least in part, on the node power consumption and a component power consumption. The system can determine a power budget for the component for a second time interval based, at least in part on the proportional component power budget.

Distributed power budgeting

A system with a local data collector that collects power management data for a subsystem. The local data collector can determine whether a first formatting associated with a first channel between the local data collector and a system power management data collector is equivalent to a second formatting associated with a second channel between the local data collector and the system power management data collector, and in response to a determination that the first formatting and second formatting are not equivalent format the power management data according to the first formatting; store the power management data formatted according to the first formatting in a first location in a memory; format the power management data according to the second formatting; and store the power management data formatted according to the second formatting in a second location the memory.

Scalable data collection for system management

A method and apparatus for adjusting a frequency of a processor is disclosed herein. In one embodiment, the method includes inhibiting one or more processor cores from exiting an idle state. The method further includes determining a number of processor cores requesting exit from the idle state and a number of non-idle processor cores. The method also includes selecting a maximum frequency for the inhibited processor cores based on the number of inhibited processor cores requesting exit from the idle state and the number of non-idle processor cores. The method includes setting the maximum frequency for the inhibited processor cores, and then uninhibiting the processor cores requesting exit from the idle state.

Management of core power state transition in a microprocessor

A system for adjusting a frequency of a processor is disclosed herein. The system includes a processor and a memory, where the memory includes a program configured to adjust a frequency of a multi-core processor. The operations include determining a total current and a temperature of the multi-core processor and estimating a leakage current for the multi-core processor. The operations also include calculating a switching current by subtracting the leakage current from the total current and calculating an effective switching capacitance based at least in part on the switching current. The operations also include calculating a workload activity factor by dividing the effective switching capacitance by a predetermined effective switching capacitance stored in vital product data, and enforcing a turbo frequency limit of the multi-core processor based on the workload activity factor.

Deterministic current based frequency optimization of processor chip

An apparatus includes a plurality of components and a plurality of component controllers. Each of the plurality of component controllers is associated with at least one component of the plurality of components. Each component controller is configured to compute a local power budget for the at least one component based, at least in part, on the power differential and the proportion of the total power consumption corresponding to the at least one component. A service processor is configured to determine failure associated with at least one component controller of the plurality of component controllers or the at least one component associated with the at least one component controller. The service processor is configured to in response to a reset threshold not being exceeded, reset the at least one component controller without interrupting operations of any components of the at least one component that have not failed.

Handling a failure in a system with distributed control of power and thermal management

A method for adjusting a frequency of a processor is disclosed herein. In one embodiment, the method includes inhibiting one or more processor cores from exiting an idle state. The method further includes determining a number of processor cores requesting exit from the idle state and a number of non-idle processor cores. The method also includes selecting a maximum frequency for the inhibited and non-idle processor cores based on the number of inhibited processor cores requesting exit from the idle state and the number of non-idle processor cores. The method includes setting the maximum frequency for both the inhibited and the non-idle processor cores, and then uninhibiting the processor cores requesting exit from the idle state.

Component power consumption is collected from each of a plurality of controllers of a node having a plurality of components. The component power consumption is provided to each of the plurality of controllers. A power differential is determined as a difference between a power cap for an apparatus and a total power consumption for the apparatus based, at least in part, on the component power consumption. A proportion of the total power consumption corresponding to the at least one component associated with the at least one component controller is determined. A local power budget is computed for the at least one component based, at least in part, on the power differential and the proportion of the total power consumption corresponding to the at least one component. A failure associated with the at least one component controller or the at least one component is determined.

A method for adjusting a frequency of a processor is disclosed herein. In one embodiment, the method includes determining a total current and a temperature of the multi-core processor and estimating a leakage current for the multi-core processor. The method also includes calculating a switching current by subtracting the leakage current from the total current. The method also includes calculating an effective switching capacitance based at least in part on the switching current. The method also includes calculating a workload activity factor by dividing the effective switching capacitance by a predetermined effective switching capacitance stored in vital product data, and enforcing a turbo frequency limit of the multi-core processor based on the workload activity factor.

It is determined that a current node power consumption for a node is greater than a node power cap that defines a limit of power consumption for the node. Responsive to the current node power consumption being greater than the node power cap and until the current node power consumption is less than the node power cap, power reduction operations are performed. The power reduction operations comprise determining a power management zone of a plurality of power management zones having a lowest priority among the power management zones and having a power cap greater than a minimum power cap for the power management zone. The power reduction operations further comprise setting the power cap for the power management zone to a value less than a prior value assigned as the power cap for the power management zone.

Computer program product and a node implementing power management by associated power management controllers based on distributed node power consumption and priority data

Method performed by an associated power management controller of a zone based on node power consumption and priority data for each of the plurality of zones

A zone power cap for a power management zone that defines a limit of power consumption for the power management zone is determined. The power management zone comprises a plurality of components, wherein the power management zone is associated with a controller. A set of one or more characteristics of a workload associated with the power management zone is determined. A component power cap for one or more of the plurality of components is set based, at least in part, on the set of one or more characteristics of the workload and the zone power cap.

Distributed power management system with plurality of power management controllers controlling zone and component power caps of respective zones by determining priority of other zones

Growing community of inventors

Zumbrota, MN, United States of America

Todd Jon Rosedahl