Google Inc.

Exaflop LLC

Xiaobo Fan

Wolf-Dietrich Weber

Luiz Andre Barroso

William Riis Hamburgen

Jimmy Clidaras

Andrew B Carlson

Taliver Heath

Mark D Hennecke

Selver Corhodzic

Chris Sadler

Taliver Brooks Heath

A method of designing a data center can include determining a design power density, determining an oversubscription ratio, and determining a spatial layout of the data center using the design power density and the oversubscription ratio. The design power density can be in units of power per unit length.

Data center design

A method for use in deploying computers into a data center includes calculating in a computer an expected peak power draw for a plurality of computers. The expected peak power draw for the plurality of computers is less than a sum of individual expected peak power draws for each computer from the plurality of computers.

Computer and data center load determination

The subject matter of this specification can be embodied in, among other things, a method that includes supplying power to a portion of a data center through a power distribution line. Utilization of a statistically significant sample of the computers is monitored, and an estimated individual power draw for each of the computers based on the utilization is calculated. An estimated total power draw is calculated for different times from the estimated individual power draws to generate a collection of estimated total power draw values for the different times. Actual power draw is monitored at the power distribution line and a collection of actual power draw values is generated. A function is fitted to pairs of actual power draw values and estimated power draw values, each pair comprising an actual draw value and an estimated draw value for the same time, and the function is then stored.

Accurate power allotment

Method of correlating power in a data center by fitting a function to a plurality of pairs of actual power draw values and estimated power draw values determined from monitored CPU utilization of a statistical sample of computers in the data center

A method of monitoring power load of a data center includes monitoring a power usage of a cluster of the data center, monitoring a power usage of a power distribution unit (PDU) of the data center, and monitoring a power usage of at least one of a rack of the data center or a computer within a rack of the data center.

Data center load monitoring for utilizing an access power amount based on a projected peak power usage and a monitored power usage

In a computer-implemented method, an electronic communication from a first computing machine is received and includes a power value and one or more priority values. The power value represents a request for a power allotment expected to be used during a predetermined time and the priority values represent priorities of tasks expected to be executed during the predetermined time. A score for the request is calculated using a scoring function and the received priority values as inputs to the scoring function. The score is compared to one or more other scores respectively associated with requests for power allotments from other computing machines and previously calculated using the scoring function following receipt of one or more electronic communications each including a power value and one or more priority values. A top-ranked score is identified and an electronic communication is sent to the associated computing machine granting the requested power allotment.

Power allotment distribution in a data center

A method of controlling power usage in a data center includes monitoring a power usage of a plurality of computers in the data center, generating a signal indicating that the power usage is within a threshold of a maximum power capacity, and in response to the signal, adjusting performance of at least one computer in the plurality of computers.

Load control in a data center

A data center includes a power distribution network having a power capacity, and a plurality of computers drawing power from the power distribution network. Each of the computers has a peak power draw. The power capacity is less than a maximum power draw defined by summing the peak power draw from each of the plurality of computers.

Powering a data center

Exemplary systems, apparatus, and methods relate to a current limiting device on each of two or more load branches that limits the current drawn by each load branch to a branch current allocation for that branch, where each of the current limiting devices is configured to automatically communicate messages to negotiate adjustments to the branch current allocations among each of the load branches while maintaining the sum of the branch allocations substantially at or below an available capacity of the supply branch. In an exemplary embodiment, a system of intelligent power modules (IPMs) may operate together to automatically negotiate capacity sharing among a number of load branches by adapting to the dynamic load conditions in order to takes improved advantage of infrastructure power handling capability without exceeding a predetermined capacity of the supply branch.

Negotiating capacity allocation among distributed current protection devices

A facility is described that includes one or more enclosures defining an interior space, a plurality of power taps, a plurality of coolant supply taps, and a plurality of coolant return taps. A flow capacity of the supply taps and a flow capacity of the return taps can be approximately equal over a local area of the interior space. The plurality of power taps, the plurality of supply taps, and the plurality of return taps can be divided into a plurality of zones, with taps of each zone are configured to be controllably coupled to a power source or a coolant source independently of the taps of other zones. The taps can be positioned along paths, and paths of the power taps can be spaced from associated proximate paths of supply and return taps by a substantially uniform distance along a substantial length of the first path.

Growing community of inventors

Sunnyvale, CA, United States of America

Xiaobo Fan