Abstract: In disclosed techniques, simulations are performed to determine data center performance under certain conditions. The simulations are dynamic and allow for changes in power demand due to temporal data center activities. In order to accommodate predicted and unpredicted fluctuations in power demand of a data center, one or more power caches are configured to supply additional power during periods of peak demand. Power caches provide supplemental power during periods of peak demand. The simulations are used for a variety of purposes, including determining the effects of power caches going offline under various conditions. Disclosed techniques can simulate the cycling of a power cache and can determine if additional configuration changes to the data center are warranted to maintain optimal health of the power caches. Thus, power scenario simulation of a data center can provide information vital to efficient operation of the data center.

Abstract: In disclosed techniques, datacenter power management uses AC and DC power sources. An AC power distribution topology within a datacenter provides one or more AC power sources to computing devices. A DC power distribution topology within the datacenter provides one or more DC power sources to computing devices. An uninterruptible power supply (UPS) is provisioned to a rack of computing devices, wherein the UPS is capable of receiving the one or more AC power sources using the AC power distribution topology and the one or more DC power sources using the DC power distribution topology. The one or more DC power sources are evaluated for energizing the DC power distribution topology. The one or more DC power sources are connected to the UPS which is provisioned to the rack of computing devices, based on the evaluating and a datacenter power requirement.

Abstract: Disclosed techniques include systems and methods for managed negotiation between a controller of a datacenter and devices receiving resources as directed by the controller in the datacenter. The controller can identify a need for reduction in power provision or any other resource, and can make offers to device(s) requesting the device consumes less power. The offer also includes compensation paid by the controller to the device in exchange for the power reduction.

Abstract: Embodiments provide techniques for datacenter power management using variable power sources. Power from the variable power sources is stored in a power cache. An optimization engine receives input criteria such as power availability from non-variable and variable power sources, as well as one or more power management goals. The optimization engine implements a dispatch strategy that dispatches stored energy from the power cache and feeds it to the datacenter, resulting in a mixture of non-variable and variable power sources used to achieve the power management goals, such as reduced power cost, increased power availability, and lowered carbon footprint for the datacenter.

Abstract: Datacenter power management using edge mediation is disclosed. A datacenter power policy is provided to a centralized power management platform. The policy oversees a datacenter topology that includes energy control blocks. Each of the energy control blocks comprises an energy control block coupled to one or more datacenter loads. A power control gateway is coupled between the centralized power management platform and the one or more of energy control blocks. The power control gateway comprises one gateway element for each of the one or more energy control blocks. Each gateway element includes a policy execution component, a sensor data exchange component, a data summarization component, and a global state analysis component. Sensor data from the energy control blocks is summarized and forwarded to the centralized power management platform. Power distribution to the one or more datacenter loads is controlled using inter-gateway element information distributed among each gateway element.

Abstract: Dynamic tiering of datacenter power for workloads is disclosed. A power capacity, including redundant power capacity and granular power capacity values within a datacenter, is determined. An outage time duration requirement for the power capacity that was determined is evaluated, where the outage time duration requirement is a number of minutes. A hold time duration requirement for the power capacity is evaluated, where the hold time duration is a number of minutes. A number of allowable occurrences of power outage for the power capacity is evaluated. A power requirement metric, based on the outage time duration requirement, the hold time duration requirement, and the number of occurrences, is calculated. A power topology within the datacenter is modified based on the power requirement metric. The modifying provides dynamic power tiering within the datacenter. The dynamic tiering includes a variable service level agreement for power within the datacenter.

Abstract: Datacenter current injection for power management is disclosed. A controller for managing power control modules and current sensing for a datacenter power infrastructure unit is provided. At least three power control modules coupled to the controller are each connected to each phase of a three-phase datacenter power grid. Current sense inputs for each phase of the datacenter power grid are coupled to the controller. At least three battery modules are coupled to a corresponding power control module. The coupling provides an energy path between a battery module and phase of the three-phase datacenter power grid. The coupling is managed by the controller. Each power control module includes a battery charging unit and a current mode grid tie inverter unit. The current mode grid tie inverter unit enables current injection into each phase of the three-phase datacenter power grid at constant voltage and in-phase AC frequency for the phase.

Abstract: Datacenter power management using current injection is disclosed. A datacenter topology that includes a utility grid feed, power caches, power control blocks, loads, and uninterruptible power supplies (UPSs) providing AC power is accessed. A supply capacity is allocated from a first UPS to the one or more loads within the datacenter, below a peak load requirement for the loads. It is also allocated based on a datacenter power policy. An AC current requirement is detected by the one or more power control blocks. The AC current requirement is detected for the one or more loads at the output network of the first UPS. AC current is injected into the output network of the first UPS by the one or more power control blocks. The AC current injection is based on the AC current requirement that was detected. The AC current injection is further based on the datacenter power policy.

Abstract: Disclosed techniques include time-varying power management within datacenters. A set of power policies for managing power within a datacenter is obtained. The set of policies varies over time. A priority is determined for a policy within the set of policies for managing the power within the datacenter. The policy within the set varies over time. A situation within the datacenter is identified where the situation matches that described in the policy within the set of policies. A power arrangement within the datacenter is modified based on the policy within the set of policies. The power arrangement within the datacenter applies to a section of the datacenter including one or more IT racks. The modifying includes powering a set of loads within the datacenter by a specific power source, changing a topology within the datacenter, powering down a portion of the datacenter, or changing a service level agreement support level.

Abstract: In disclosed techniques, augmented power control within a datacenter uses predictive modeling. A power usage by a first data rack within a datacenter is measured, using one or more processors, over a first period of time. A predicted power usage by the first data rack over a second period of time is generated on a computing device, wherein the second period of time is subsequent to the first period of time. A power correlation model is calculated that correlates a power prediction to an actual power usage. The predicted power usage for the second period of time is refined, based on the predicted power usage and the power correlation model. The refining is accomplished using a power prediction model comprising the predicted power usage and the power correlation model. Datacenter power structure is configured, based on the refined power usage prediction.

Abstract: In disclosed techniques, secure communication initiation and execution is used for datacenter power control. Information relating to power control is encrypted for inclusion in a first data payload. The first data payload is used for datacenter power infrastructure control. The first data payload is sent from a first component within a datacenter to a second component within the datacenter. The first component and the second component enable power infrastructure power control of the datacenter. The datacenter power infrastructure control is modified based on decryption of the first data payload by the second component within the datacenter. The first data payload provides for intelligent power control within the datacenter. Modifying the datacenter power infrastructure control dynamically changes power control within the datacenter. The changing power control changes power policies within the datacenter. The first component is authenticated using the first data payload. Encryption includes obscurity.

Abstract: In disclosed techniques, secure communication initiation and execution is used for datacenter power control. Information relating to power control is encrypted for inclusion in a first data payload. The first data payload is used for datacenter power infrastructure control. The first data payload is sent from a first component within a datacenter to a second component within the datacenter. The first component and the second component enable power infrastructure power control of the datacenter. The datacenter power infrastructure control is modified based on decryption of the first data payload by the second component within the datacenter. The first data payload provides for intelligent power control within the datacenter. Modifying the datacenter power infrastructure control dynamically changes power control within the datacenter. The changing power control changes power policies within the datacenter. The first component is authenticated using the first data payload. Encryption includes obscurity.

Abstract: Disclosed techniques include power manipulation using power caches within a datacenter. The power caches store power for later use when power demand exceeds the power available. A power source is coupled across a plurality of data racks to a power load within a datacenter. A power cache is coupled to the power load. Power for the power load is provided from the power source. The power cache is enabled to provide power to the power load when power requirements of the power load exceed a power threshold. The power threshold is determined by a set of power policies for the datacenter. The enabling the power cache to provide power to the power load provides power bursting capability.

Abstract: Disclosed techniques enable power manipulation within a data center based on power policies. Power policies are used to establish behavior based on limits for power consumption, power generation, and other criteria. Rules are established to provide fractional power sharing and control under certain conditions. Power caches are used to supplement power sources under conditions of peak power requirements. When power requirements are below a threshold set forth in a power policy, the power caches are replenished.

Abstract: Techniques for datacenter power management using dynamic redundancy are disclosed. A power control switch is configured to selectively apply power to one or two power cords of a dual-corded electronic apparatus. When the power control switch energizes both power cords, the electronic apparatus operates in 2N redundancy. When the power control switch energizes only one of the power cords, the electronic apparatus operates in 1N redundancy. The power control switch is configured to dynamically change the redundancy mode based on service level agreement (SLA) criteria, power policies, power supply and demand, and environmental factors.

Abstract: Techniques for datacenter power management using dynamic redundancy are disclosed. A power control switch is configured to selectively apply power to one or two power cords of a dual-corded electronic apparatus. When the power control switch energizes both power cords, the electronic apparatus operates in 2N redundancy. When the power control switch energizes only one of the power cords, the electronic apparatus operates in 1N redundancy. The power control switch is configured to dynamically change the redundancy mode based on service level agreement (SLA) criteria, power policies, power supply and demand, and environmental factors.

Abstract: Multi-level data center using consolidated power control is disclosed. One or more controllers communicate with one or more power supplies and sensors to adapt to dynamic load requirements and to distribute DC-power efficiently across multiple IT racks. The distributed DC-power includes one or more DC-voltages. Batteries can be temporarily switched into a power supply circuit to supplement the power supply during spikes in power load demand while a controller reconfigures power supplies to meet the new demand. One or more DC-to-AC converters handle legacy power loads. The DC-to-AC converters are modular and are paralleled for redundancy. The DC-to-AC converters are sourced from the one or more AC-to-DC converters. The AC power is synchronized for correct power flow control. The one or more controllers communicate with the one or more AC-power supplies to dynamically allocate AC-power from the DC-to-AC converters to the multiple legacy IT racks.

Abstract: Dynamic tiering of datacenter power for workloads is disclosed. A power capacity, including redundant power capacity and granular power capacity values within a datacenter, is determined. An outage time duration requirement for the power capacity that was determined is evaluated, where the outage time duration requirement is a number of minutes. A hold time duration requirement for the power capacity is evaluated, where the hold time duration is a number of minutes. A number of allowable occurrences of power outage for the power capacity is evaluated. A power requirement metric, based on the outage time duration requirement, the hold time duration requirement, and the number of occurrences, is calculated. A power topology within the datacenter is modified based on the power requirement metric. The modifying provides dynamic power tiering within the datacenter. The dynamic tiering includes a variable service level agreement for power within the datacenter.

Abstract: Disclosed techniques include time-varying power management within datacenters. A set of power policies for managing power within a datacenter is obtained. The set of policies varies over time. A priority is determined for a policy within the set of policies for managing the power within the datacenter. The policy within the set varies over time. A situation within the datacenter is identified where the situation matches that described in the policy within the set of policies. A power arrangement within the datacenter is modified based on the policy within the set of policies. The power arrangement within the datacenter applies to a section of the datacenter including one or more IT racks. The modifying includes powering a set of loads within the datacenter by a specific power source, changing a topology within the datacenter, powering down a portion of the datacenter, or changing a service level agreement support level.

Abstract: In disclosed techniques, secure communication initiation and execution is used for datacenter power control. Information relating to power control is encrypted for inclusion in a first data payload. The first data payload is used for datacenter power infrastructure control. The first data payload is sent from a first component within a datacenter to a second component within the datacenter. The first component and the second component enable power infrastructure power control of the datacenter. The datacenter power infrastructure control is modified based on decryption of the first data payload by the second component within the datacenter. The first data payload provides for intelligent power control within the datacenter. Modifying the datacenter power infrastructure control dynamically changes power control within the datacenter. The changing power control changes power policies within the datacenter. The first component is authenticated using the first data payload. Encryption includes obscurity.