Abstract: A self-powered processing device comprises both a processing device and a power generator that are physically, electrically, and thermally coupled to one another. The power generator can be a fuel cell that can be manufactured from materials that can also support processing circuitry, such as silicon-based materials. A thermal coupling between the power generator and the processing device can include a thermoelectric either generating electrical power from the temperature differential or consuming electrical power to generate a temperature differential. A computing device with self-powered processing devices also includes energy storage devices to store excess energy produced by the self-powered processing device and provide it back during times of need. The self-powered processing device comprises either a wireless or wired network connection, the latter being connectable to a slot on a backplane that can aggregate multiple self-powered processing devices and provide fuel delivery paths for them.

Abstract: Water condensate is captured from the exhaust of a generator utilized to provide power to a data center, and the captured water is then utilized for data center purposes such as adiabatic cooling. The exhaust of electrical power generators is passed through a condenser to obtain water condensate from such exhaust. The water condensate is stored in water storage units and is utilized to provide supplemental cooling to the data center. Sporadic usage of water can enable the water storage to be refilled between uses, since water condensate can be obtained from exhaust almost continuously. The level of water is monitored and the level of processing performed by the data center is adjusted to avoid emptying such water storage units. Historical climatological data is utilized to estimate the water required. Additionally, short and long-range weather forecasts can be optionally taken into account.

Abstract: Computing devices receive power from multiple fuel cells, consuming natural gas and outputting electrical energy natively consumable by the computing devices. The fuel cells are sized to provide power to a set of computing devices, such as a rack thereof. The computing devices of a failed fuel cell can receive power from adjacent fuel cells. Additionally, the fuel cells and computing devices are positioned to realize thermal symbiotic efficiencies. Controllers instruct the computing devices to deactivate or throttle down power consuming functions during instances where the power consumption demand is increasing faster than the power being sourced by fuel cells, and instruct the computing devices to activate or throttle up power consuming functions during instances where the power consumption demand is decreasing faster than the power being sourced by the fuel cells. Supplemental power sources, supplementing the fuel cells' inability to quickly change power output, are not required.

Abstract: Gas supply pressure spikes are absorbed and leveled-out by a gas supply shock absorber comprising gas storage, which is charged during positive pressure spikes and utilized during negative pressure spikes. The gas supply shock absorber also comprises pressure sensing and regulating valves, which direct positive pressure spikes to the gas storage and draw gas from storage during negative pressure spikes. A backflow preventer limits shock absorption to co-located equipment, but gas supply shock absorbers operate in aggregate to create additional demand during positive pressure spikes and reduced demand during negative pressure spikes. If the gas storage has sufficient gas, a co-located data center utilizes such gas for increased electrical power generation during increased processing activity, which can be requested or generated. Conversely, if the gas storage has insufficient gas, and a negative pressure spike occurs, the data center throttles down or offloads processing.

Abstract: A “data plant” accepts power-generation-capable raw materials and outputs processed data. The processed data can be delivered to consumers more efficiently than other forms of power transfer, including power transfer through electricity, steam, physical motion, and the like. Consequently, data plants can be located where power-generation-capable raw materials can be obtained inexpensively, for free, or where power-generation-capable raw materials are waste products for which the operator of the data plants can be compensated for processing. Self-powered data plants need not even be continuously fed with power-generation-capable raw materials and, if such data plants receive and output data via wireless communications, the self-powered data plants can require no physical connection or attachment at all. For example, a single piece of silicon comprising a silicon solar cell that generates electrical power and silicon circuitry that consumes it to perform data processing can be a silicon self-powered data plant.

Abstract: A power supply system for a data center includes a cooling circuit, an electrochemical power generator, a sensor, and a processor. The cooling circuit includes a fluid configured to receive heat energy generated by a server located in the data center. The electrochemical power generator is configured to receive and/or generate the fluid of the cooling circuit and to generate electrical energy for the server using the fluid. The sensor is configured to obtain data regarding the server. The processor is configured to control an amount of heat energy transferred from the server to the fluid based on the data.

Abstract: A data center is cooled through hydronic convection mechanisms, geothermal mechanisms or combinations thereof. The individual computing devices of such a data center are cooled through a thermally conductive interface with a liquid. The liquid's container can extend to a cooling apparatus located physically above such computing devices to provide hydronic convection cooling, or it can extend into the earth, either in the form of a heat pipe, or in the form of conduits through which the liquid is actively pumped. The hydronic convection cooling and geothermal heat pipe cooling operate via temperature differentials and consume no external electrical power. Geothermal cooling avoids heat soak by utilizing multiple different sets of conduits extending into the earth, where at least some of those sets of conduits are not utilized for a period of time. Combinations of hydronic convection mechanisms and geothermal cooling can also be utilized.

Abstract: A “Cascading Startup Controller” provides various techniques for quickly and efficiently initializing grids of interconnected fuel cells. In general, the Cascading Startup Controller dynamically controls heat exchange between fuel cells in the grid to produce a cascading startup of the fuel cell grid via an expanding pattern of excess thermal energy routing from hotter fuel cell stacks to cooler fuel cell stacks. This expanding pattern of excess thermal energy routing is dynamically controlled via automated valves of a heat exchange grid coupled to the fuel cell grid to decrease a total startup time for fuel cell stacks in the grid. Additional excess heat beyond that used to heat fuel cells to operational temperatures is then made available for a variety of purposes, including, but not limited to, preheating gas or other fuel for use by the fuel cells, local or community-based heating systems, heat-based energy cogeneration systems, etc.

Abstract: A “data plant” accepts power-generation-capable raw materials and outputs processed data. The processed data can be delivered to consumers more efficiently than other forms of power transfer, including power transfer through electricity, steam, physical motion, and the like. Consequently, data plants can be located where power-generation-capable raw materials can be obtained inexpensively, for free, or where power-generation-capable raw materials are waste products for which the operator of the data plants can be compensated for processing. Self-powered data plants need not even be continuously fed with power-generation-capable raw materials and, if such data plants receive and output data via wireless communications, the self-powered data plants can require no physical connection or attachment at all. For example, a single piece of silicon comprising a silicon solar cell that generates electrical power and silicon circuitry that consumes it to perform data processing can be a silicon self-powered data plant.

Abstract: Technology for concurrently powering equipment from multiple power sources, and the control thereof is disclosed. One example implementation of the technology includes a first power supply that powers equipment from a first power source and a second power supply that also powers the equipment from a second power source while the equipment is being powered by the first power supply. A target direct current (DC) output voltage of at least one of the power supplies is changed, thereby changing a ratio of the power being drawn from the first power supply to the power being drawn from the second power supply.

Abstract: An analysis system is described herein for detecting anomalies within an environment based on a consideration of resources supplied to, and then used by, resource consumption devices within the environment. For instance, in one implementation, the analysis system may detect leaks of natural gas in a data processing environment based on a consideration of discrepancies in the amounts of gas supplied to the resource consumption devices, relative to the amounts of energy produced by the resource consumption devices, as a result of the use of the gas. In addition, or alternatively, the analysis system may detect abnormal degradation of the resource consumption devices within the data processing environment, such as its fuel cells or generators. That degradation may be attributable to intrinsic failures associated with the fuel cells or generators, and/or to the nature of the loads that have been applied to the fuel cells or generators.

Abstract: A power supply is described herein which provides power to a load, such as a load including one or more computing devices. The power supply uses a slow-response power source (such as a fuel-driven mechanism) to handle a slow-moving component of the demand level presented by the load, and uses a fast-response power source (such as a battery or a capacitor, etc.) to handle a fast-moving component of the demand level. By virtue of this approach, the power supply can manage the load level as it appears to the slow-response power source, allowing, in turn, the slow-response power source to service even fast-changing loads—a task which it could not otherwise perform due to its native limitations.

Abstract: The power consumption of a computing device is inferred from the utilization rates of individual components of the computing device and a utilization-to-power-consumption transfer function that was derived by benchmarking that, or an analogous, computing device. The inferred power consumption of a computing device is aggregated to infer the power consumption of various groups and super-groups of computing devices. The historical power consumption of computing devices is inferred based on the utilization rates of individual components of the computing devices at relevant times in the past. Historical power consumption is used to derive a power consumption profile of a computing device and the inferred current power consumption of such a computing device is compared to such a power consumption profile, and to the historical power consumption, to identify deviations therefrom, which can provide proactive detection of potential hardware faults, software glitches, or other errors.

Abstract: A self-powered processing device comprises both a processing device and a power generator that are physically, electrically, and thermally coupled to one another. The power generator can be a fuel cell that can be manufactured from materials that can also support processing circuitry, such as silicon-based materials. A thermal coupling between the power generator and the processing device can include a thermoelectric either generating electrical power from the temperature differential or consuming electrical power to generate a temperature differential. A computing device with self-powered processing devices also includes energy storage devices to store excess energy produced by the self-powered processing device and provide it back during times of need. The self-powered processing device comprises either a wireless or wired network connection, the latter being connectable to a slot on a backplane that can aggregate multiple self-powered processing devices and provide fuel delivery paths for them.

Abstract: A data center is cooled through hydronic convection mechanisms, geothermal mechanisms or combinations thereof. The individual computing devices of such a data center are cooled through a thermally conductive interface with a liquid. The liquid's container can extend to a cooling apparatus located physically above such computing devices to provide hydronic convection cooling, or it can extend into the earth, either in the form of a heat pipe, or in the form of conduits through which the liquid is actively pumped. The hydronic convection cooling and geothermal heat pipe cooling operate via temperature differentials and consume no external electrical power. Geothermal cooling avoids heat soak by utilizing multiple different sets of conduits extending into the earth, where at least some of those sets of conduits are not utilized for a period of time. Combinations of hydronic convection mechanisms and geothermal cooling can also be utilized.

Abstract: Computing devices receive power from multiple fuel cells, consuming natural gas and outputting electrical energy natively consumable by the computing devices. The fuel cells are sized to provide power to a set of computing devices, such as a rack thereof. The computing devices of a failed fuel cell can receive power from adjacent fuel cells. Additionally, the fuel cells and computing devices are positioned to realize thermal symbiotic efficiencies. Controllers instruct the computing devices to deactivate or throttle down power consuming functions during instances where the power consumption demand is increasing faster than the power being sourced by fuel cells, and instruct the computing devices to activate or throttle up power consuming functions during instances where the power consumption demand is decreasing faster than the power being sourced by the fuel cells. Supplemental power sources, supplementing the fuel cells' inability to quickly change power output, are not required.

Abstract: Gas supply pressure spikes are absorbed and leveled-out by a gas supply shock absorber comprising gas storage, which is charged during positive pressure spikes and utilized during negative pressure spikes. The gas supply shock absorber also comprises pressure sensing and regulating valves, which direct positive pressure spikes to the gas storage and draw gas from storage during negative pressure spikes. A backflow preventer limits shock absorption to co-located equipment, but gas supply shock absorbers operate in aggregate to create additional demand during positive pressure spikes and reduced demand during negative pressure spikes. If the gas storage has sufficient gas, a co-located data center utilizes such gas for increased electrical power generation during increased processing activity, which can be requested or generated. Conversely, if the gas storage has insufficient gas, and a negative pressure spike occurs, the data center throttles down or offloads processing.

Abstract: Water for data center uses is generated from the ambient air by a water generator that cools air below its dew point, thereby causing the water in such air to precipitate out. The water generator is powered by renewable energy sources that provide a sufficient amount of energy over an extended period of time, despite temporary interruptions. Heated air from the data center is exhausted so as to absorb moisture from the ambient air, with such heated air being capable of holding a greater amount of moisture, and is then directed through the water generator, thereby enabling the water generator to generate a greater amount of water. The level of water available is monitored and the water generator utilizes on-demand power sources to generate a greater amount of water so as to avoid emptying water storage units.

Abstract: Water condensate is captured from the exhaust of a generator utilized to provide power to a data center, and the captured water is then utilized for data center purposes such as adiabatic cooling. The exhaust of electrical power generators is passed through a condenser to obtain water condensate from such exhaust. The water condensate is stored in water storage units and is utilized to provide supplemental cooling to the data center. Sporadic usage of water can enable the water storage to be refilled between uses, since water condensate can be obtained from exhaust almost continuously. The level of water is monitored and the level of processing performed by the data center is adjusted to avoid emptying such water storage units. Historical climatological data is utilized to estimate the water required. Additionally, short and long-range weather forecasts can be optionally taken into account.

Abstract: The power consumption of a computing device is inferred from the utilization rates of individual components of the computing device and a utilization-to-power-consumption transfer function that was derived by benchmarking that, or an analogous, computing device. The inferred power consumption of a computing device is aggregated to infer the power consumption of various groups and super-groups of computing devices. The historical power consumption of computing devices is inferred based on the utilization rates of individual components of the computing devices at relevant times in the past. Historical power consumption is used to derive a power consumption profile of a computing device and the inferred current power consumption of such a computing device is compared to such a power consumption profile, and to the historical power consumption, to identify deviations therefrom, which can provide proactive detection of potential hardware faults, software glitches, or other errors.