Abstract: Systems and methods for machine learning-based power capping and virtual machine placement in cloud platforms are disclosed. A method includes applying a machine learning model to predict whether a request for deployment of a virtual machine corresponds to deployment of a user-facing (UF) virtual machine or a non-user-facing (NUF) virtual machine. The method further includes sorting a list of candidate servers based on both a chassis score and a server score for each server to determine a ranked list of the candidate servers, where the server score depends at least on whether the request for the deployment of the virtual machine is determined to be a request for a deployment of a UF virtual machine or a request for a deployment of an NUF virtual machine. The method further includes deploying the virtual machine to a server with highest rank among the ranked list of the candidate servers.

Abstract: The discussion relates to disaggregated computing. One example can monitor multiple two-phase liquid immersion tanks. Individual two-phase liquid immersion tanks can contain multiple components of a single type of component type. The example can receive requests for virtual machines and allocate sets of components from individual two-phase liquid immersion tanks to work together to support the virtual machines requests.

Abstract: Systems and methods for machine learning-based power capping and virtual machine placement in cloud platforms are disclosed. A method includes applying a machine learning model to predict whether a request for deployment of a virtual machine corresponds to deployment of a user-facing (UF) virtual machine or a non-user-facing (NUF) virtual machine. The method further includes sorting a list of candidate servers based on both a chassis score and a server score for each server to determine a ranked list of the candidate servers, where the server score depends at least on whether the request for the deployment of the virtual machine is determined to be a request for a deployment of a UF virtual machine or a request for a deployment of an NUF virtual machine. The method further includes deploying the virtual machine to a server with highest rank among the ranked list of the candidate servers.

Abstract: A thermal management system for a computing device includes an immersion tank with a cooling fluid therein, a computing device positioned in the cooling fluid in the immersion tank, and a thermal block positioned in the cooling fluid in the immersion tank. The computing device heats the cooling fluid, and the thermal block is configured to receive heat from the cooling fluid. The thermal block includes a fluid management feature to direct flow of the cooling fluid relative to the thermal block and computing device.

Abstract: The present disclosure relates to systems, methods, and computer readable media for enabling server devices to utilize a higher percentage of power resources while maintaining sufficient availability of power resources of a datacenter or other collection of server devices. For example, systems disclosed herein determine and implement power shaving actions based on virtual machine metadata and in accordance with a power shaving policy to facilitate a significantly higher utilization of power resources on a datacenter during normal operation as well as within periods of limited power capacity on various server devices. Systems described herein provide more efficient utilization of power resources while maintaining service availability guarantees for a variety of virtual machines hosted by servers of the datacenter.

Abstract: A method of controlling energy in a datacenter includes receiving a fuel cell operating percentage of an operating capacity of the fuel cell, receiving a fuel cell exhaust temperature, receiving a hot aisle air temperature from a hot aisle of a server computer, determining a temperature delta between the hot aisle air temperature and the fuel cell exhaust temperature, and then allocating virtual machine placements to change a server user percentage relative to a server user capacity percentage target value to optimize the fuel cell operating percentage relative to the fuel cell efficiency target value, the temperature delta relative to the thermoelectric generator efficiency target value, and the server user percentage relative to the server user capacity percentage target value.

Abstract: A method of controlling energy in a datacenter includes receiving a fuel cell operating percentage of an operating capacity of the fuel cell, receiving a fuel cell exhaust temperature, receiving a hot aisle air temperature from a hot aisle of a server computer, determining a temperature delta between the hot aisle air temperature and the fuel cell exhaust temperature, and then allocating virtual machine placements to change a server user percentage relative to a server user capacity percentage target value to optimize the fuel cell operating percentage relative to the fuel cell efficiency target value, the temperature delta relative to the thermoelectric generator efficiency target value, and the server user percentage relative to the server user capacity percentage target value.

Abstract: Systems and methods for machine learning-based power capping and virtual machine placement in cloud platforms are disclosed. A method includes applying a machine learning model to predict whether a request for deployment of a virtual machine corresponds to deployment of a user-facing (UF) virtual machine or a non-user-facing (NUF) virtual machine. The method further includes sorting a list of candidate servers based on both a chassis score and a server score for each server to determine a ranked list of the candidate servers, where the server score depends at least on whether the request for the deployment of the virtual machine is determined to be a request for a deployment of a UF virtual machine or a request for a deployment of an NUF virtual machine. The method further includes deploying the virtual machine to a server with highest rank among the ranked list of the candidate servers.

Abstract: Techniques for mitigating loss of vaporized working fluid in a two-phase immersion cooling system may be implemented using one or more supplemental condensers that facilitate condensation of vaporized working fluid when the immersion tank is open, and one or more vapor collection points that are in fluid communication with at least one supplemental condenser. One or more fluid displacement devices may be configured to create suction pressure at the one or more vapor collection points. One or more vents may be positioned in the door. The one or more vents may be configured to permit movement of air from outside the immersion tank into an interior portion of the immersion tank without permitting loss of vaporized working fluid. A directional blowing device may be configured to blow a gaseous substance against a computing device in a downward direction as the computing device is being pulled upward out of the immersion tank.

Abstract: A method for facilitating increased utilization of a data center includes receiving information about availability of components in a data center's electrical infrastructure and about power consumption of servers in the data center. The method may also include detecting that the power consumption of the servers in the data center exceeds a reduced total capacity of the electrical infrastructure. The reduced total capacity may be caused by unavailability of at least one component in the data center's electrical infrastructure. The method may also include causing power management to be performed to reduce the power consumption of the servers so that the power consumption of the servers does not exceed the reduced total capacity of the electrical infrastructure of the data center.

Abstract: A system estimates a future demand for resources over a first time period to build a datacenter. The resources include construction resources to construct the datacenter and computing resources for the datacenter. The system generates, based on inputs including the future demand, a first schedule of resources to build a first stage of the datacenter over the first time period. The first schedule includes fewer resources than an amount of resources capable of fulfilling the future demand. The system determines a probability of the first schedule not fulfilling the future demand and determines a risk associated with the first schedule based on the probability. The system identifies procedures to employ to mitigate the risk, including oversubscribing resources in the first schedule, lending a portion of the future demand to a peer datacenter, and leasing an additional datacenter, where the first schedule and the procedures minimize cost of building the datacenter.