Abstract: A system and related method balance phase power in a power distribution system. The method comprises performing the following operations dynamically and repeatedly, during operation of subcomponents receiving power from a plurality of redundant intelligent PDUs. Input phases of the intelligent PDUs are monitored for power values associated with the input phases that provide power to the subcomponents through the PDUs. A target per-phase power value is determined. When the input phases are not balanced, an input phase at an input port is redirected to an output port of at least one of the PDUs using a switch to bring the not-balanced phases closer to being balanced phases by reducing a difference of phase powers to the target per-phase power value. The method then comprises performing load shifting on the power supply units (PSUs) associated with the redundant PDUs to improve an overall phase power balance across all redundant PDUs.

Abstract: Disclosed embodiments provide methods, systems, and computer program products for implementing enhanced phase power load balancing of a set of systems (e.g., the set of systems of a data center) using a subset of the systems. Disclosed embodiments extract phase currents of a three-phase power source of the set of systems, and determine, based on extracted phase currents for the set of systems, input phase power load balancing for the subset of the systems. Disclosed embodiments alter the current input phase power balancing of the subset, unbalancing the input phase power load for the subset of the systems to provide enhanced phase power load balancing of the set of systems. Altering the current input phase power balancing for the subset of the systems can be implemented by power shifting between power supply units (PSUs), phase rotations, or input phase switching of power distribution unit (PDU) switchable output ports.

Abstract: Disclosed embodiments provide methods, systems, and computer program products for implementing intelligent power shifting in an uninterruptable power supply (UPS) system for enhanced system efficiency. Disclosed embodiments implement intelligent power shifting to decrease UPS power consumption during normal system operation in the presence of a main power signal, increasing system efficiency, while providing redundancy in the absence or a disruption of the main power signal. Power shifting is based on an intelligent algorithm for controlling current through power supply units (PSUs) of the UPS system to shift a substantial portion of the main power signal through PSUs that are connected to power distribution units (PDU) not powered through the UPS, reducing power consumption by the UPS.

Abstract: A computer-implemented apparatus and related method control a dynamic phase rotation unit of a power distribution unit (PDU). The method comprises using a processor of a system controller for identifying a PDU group of two or more redundant PDUs, receiving input parameters from an input port of each of the PDUs in the PDU group, and comparing the input parameters from of the PDU group. The method further comprises conditioned upon a determination of the following both being true: 1) the input parameters are within a predefined threshold of similarity; and 2) no uninterruptable power supply (UPS) is present, then, responsive to the determination sending, via a network, a phase rotation command to a phase rotation controller to change the phase rotation switches on at least one of the PDUs in the PDU group. Otherwise, responsive to the determination, not sending the phase rotation command.

Abstract: Systems and techniques that facilitate simultaneous execution of multiple quantum jobs are provided. For example, embodiments described herein can comprise a system, which can comprise a memory that can store computer executable components. The system can also comprise a processor, operably coupled to the memory that can execute the computer executable components stored in memory. The computer executable components can comprise a scheduling component that determines a first set of qubits to execute a first quantum job on a quantum processor and a second set of qubits to execute a second quantum job, wherein the second set of qubits does not include the first set of qubits or a first set of idle qubits; and an isolation component that determines the first set of idle qubits, wherein the first set of idle qubits isolates the first set of qubits from crosstalk of other operations on the quantum processor.

Abstract: A method, system, and computer program product for optimizing development of a quantum circuit or a quantum model on a development quantum system for later use on a production quantum system. Operational and performance data are obtained from the development quantum system and the production quantum system. A noise fingerprint for the development quantum system and a noise fingerprint for the production quantum system are then generated based on the obtained operational and performance data. A noise fingerprint refers to the main features of the quantum noise sources affecting a quantum device, such as a quantum system. Parameters of the development quantum system may then be continually adjusted until the difference between the noise fingerprint for the development quantum system and the noise fingerprint for the production quantum system is below a threshold value.

Abstract: An apparatus for a fire suppressant system on a server rack includes an integrated battery feature, a manifold, a conduit, and a control card, where the integrated battery feature includes a plurality of battery cells in an enclosure. A first end of the conduit coupled to a control valve on the manifold and a second end of the conduit coupled to the integrated battery feature. The control card configured to open the control valve on the manifold, where the control valve is configured to release a fire suppressant into the enclosure of the integrated battery feature. In one embodiment, the fire suppressant is contained within a pressurized fire suppressant reservoir mounted on the server rack. In another embodiment, the fire suppressant is a cooling fluid diverted from a radiator cooling unit on the server rack.

Abstract: Embodiments herein describe a switchable redundant voltage regulator (referred to herein as a redundant phase) that can be attached to multiple different power domains. Instead of providing a redundant phase for every load or voltage domain in a computing system, the embodiments herein describe sharing redundant phases between voltage domains. Thus, if a primary phase serving any one of the phase fails, the switch can be controlled to connect the redundant phase to the corresponding voltage domain. In this manner, the redundant phase is a shared phase for a plurality of voltage domains.

Abstract: Method and apparatus for system power supply. A power demand of a system having a first power supply unit (PSU) and a second power supply unit (PSU) is monitored. An efficiency profile of the first PSU is accessed. An efficiency profile of the second PSU is accessed. A load split between the first and second PSUs is generated based on the efficiency profiles of the first and second PSUs and the power demand to increase a power efficiency of the system.

Abstract: Some implementations described herein provide apparatuses and techniques for correcting a network disruption during a virtual meeting. In response to detecting audio and/or video input to a first device, audio and/or video data is recorded and stored in a temporary buffer of the first device. After the network disruption, and using time stamps, a network disruption server may request the audio and/or video data from the first device and transmit the audio and/or video data to a second device at an accelerated rate.

Abstract: A computer-implemented apparatus and related method control a dynamic phase rotation unit of a power distribution unit (PDU). The method comprises using a processor of a system controller for identifying a PDU group of two or more redundant PDUs, receiving input parameters from an input port of each of the PDUs in the PDU group, and comparing the input parameters from of the PDU group. The method further comprises conditioned upon a determination of the following both being true: 1) the input parameters are within a predefined threshold of similarity; and 2) no uninterruptable power supply (UPS) is present, then, responsive to the determination sending, via a network, a phase rotation command to a phase rotation controller to change the phase rotation switches on at least one of the PDUs in the PDU group. Otherwise, responsive to the determination, not sending the phase rotation command.

Abstract: A method, system, and computer program product are provided for monitoring neuron growth within a lattice and stopping a spread of an infection. An infection visual recognition module monitors and captures images of the lattice during the neuron growth. The presence of the infection in the neurons within the lattice is identified. The identifying includes performing, via a deep neural network, visual recognition on the captured images to identify the presence of the infection. In response to the identifying, applying a laser to the infected area of the lattice with sufficient energy for stopping the infection. The lattice is flushed to remove chemical byproducts and dead cells resulting from the laser application.

Abstract: A method, system, and computer program product are provided. Fan speed is determined for one or more air conditioning unit (ACU) fans and stored to computer memory in response to detecting a request for authorized entrance to the modular data center. ACU fans are adjusted in preparation to deploying retractable baffles that, when deployed, form a temporary vestibule within a space of the modular data center. The retractable baffles are deployed to form the temporary vestibule within the space of the modular data center. Computing resources in the modular data center are shielded from contaminant air that enters the temporary vestibule when the external door is open. The external door is unlocked to allow user entrance to the temporary vestibule. When the external door closes, air is purged from the temporary vestibule. Retractable baffles are retracted. The fan speeds are restored from the computer memory.

Abstract: Some implementations described herein provide apparatuses and techniques for correcting a network disruption during a virtual meeting. In response to detecting audio and/or video input to a first device, audio and/or video data is recorded and stored in a temporary buffer of the first device. After the network disruption, and using time stamps, a network disruption server may request the audio and/or video data from the first device and transmit the audio and/or video data to a second device at an accelerated rate.

Abstract: Described is an edge container of an edge data center that includes information technology (IT) racks holding IT equipment. Furthermore, the edge container includes at least one air conditioning unit configured to provide cool air to an aisle which enters an inlet side of the IT racks. Additionally, the edge container includes an economizer that includes an internal and an external heat exchanger. The economizer is then activated to increase a transfer of heat/cooling between an internal environment of the edge container and an external environment in response to a reading of an environmental condition being below a temperature of exhaust air in a hot aisle of the edge container which exits an outlet side of the IT racks. If, however, the reading of the external environmental condition is not below the temperature of the exhaust air in the hot aisle of the edge container, then the economizer is deactivated.

Abstract: A method, computer program product, and computer system are provided for speech recognition-based task organization. Voice data corresponding to a task for a user to perform is received from the user. The task is identified from the received voice data through natural language processing and is classified as a parent task or a child task. The identified task is compared to prior tasks stored within a historical task database. An amount of time needed to complete the identified task is determined based on the compared prior tasks. Actions of the user are tracked and a task progress completion value is determined based on the identified amount of time.

Abstract: A method, system, and computer program product for handling black swan events on a quantum computing device. Sensor data from an environment of the quantum computing device is captured and compared to historical sensor data of the environment of the quantum computing device. A black swan event is detected if the difference between the captured sensor data and the historical sensor data exceeds a threshold value. Upon detecting a black swan event, such as during the time that the quantum processor is being utilized, a machine learning model is executed to identify the action to be performed to handle the black swan event. The machine learning model identifies such an action based on identifying a neuron of a self-organizing map that most closely matches the captured sensor data, and then identifying which of the clusters of data within the identified neuron is closest to the captured sensor data.

Abstract: A non-sealed apparatus for batteries with a containment structure for gasses that includes a battery enclosure with an upper portion and a lower portion. The apparatus further includes a plurality of batteries disposed in a lower portion of the battery enclosure. The apparatus has at least one battery containment structure disposed in an upper portion of the battery enclosure, which is configured to move downward thereby sealing the plurality of batteries upon a condition. Further, a transfer channel assembly is provided within the battery enclosure and configured to transfer and/or contain gases emitted by the plurality of batteries upon the condition. The apparatus also has at least one fan configured to circulate the air in and out of the battery enclosure through the first perforated opening and the second perforated opening.

Abstract: Some implementations described herein provide apparatuses and techniques for correcting a network disruption during a virtual meeting. In response to detecting audio and/or video input to a first device, audio and/or video data is recorded and stored in a temporary buffer of the first device. After the network disruption, and using time stamps, a network disruption server may request the audio and/or video data from the first device and transmit the audio and/or video data to a second device at an accelerated rate.

Abstract: Machine learning-based part selection in relation to one or more end use environmental conditions is provided. The process includes training a machine learning model to facilitate evaluation of a part for use in a product based on an environmental condition. Further, the process includes receiving measurement data for the part, and establishing a score for the part by comparing the measurement data for the part to a specification for the part. In addition, the method includes using the machine learning model and the established score for the part in determining whether to use the part in the product based on the environmental condition.