Abstract: The subject technology runs a compiled neural network (NN) model on a particular processor with multiple priority queues for executing different processes, the compiled NN model being assigned to a particular priority queue, and the compiled NN model includes context switch instructions that were previously inserted into a neural network (NN) model from which the compiled NN model was compiled. The subject technology determines that a particular context switch instruction has been executed by the particular processor. The subject technology determines that a different process is waiting to be executed, the different process being assigned to a different priority queue and the different process being a higher priority process than the running compiled NN model. In response to executing the particular context switch instruction, the subject technology performs a context switch to the different process assigned to the different priority queue when the different process is waiting to be executed.

Abstract: A platform-as-a-service infrastructure and application lifecycle manager is configured to implement a common services model to deploy selected services from a common set of services to service domains hosted on multiple different cloud platforms by abstracting dependence on availability of various additional supporting services, such as services that are platform-specific. The platform-as-a-service infrastructure and application lifecycle manager may also manage a lifecycle of available services, such as managing upgrades and/or patches to services.

Abstract: A system for resolving exceptions in requests determines that a request comprises an exception. The exception impeded processing the request to be granted or denied. The system determines a type of exception that indicates whether the exception is incomplete information, incorrect information, or previously-unknown information. The system generates a block in a blockchain network. Based on the stored details in the block, the type of exception is identified. Accordingly, the block is segregated for exception processing based on its exception type. The system stores the exception in the block. The system compares the request with user information previously provided by the user. The system determines a similarity score between the request and the user information. The system compares the similarity score with a threshold percentage. In response to determining that the similarity score exceeds the threshold percentage, the system determines that the exception can be resolved and resolves the exception.

Abstract: A process for facilitating autoscaling in a stateful system is described herein. In embodiments, a set of metrics associated with a set of components of a stateful service is obtained. The set of metrics may generally indicate a utilization or a load of each of the components of the set of components (e.g., message managers and/or data stores). Thereafter, it is determined to initiate a scaling event at the stateful service in association with the set of components of the stateful service based on at least a portion of the set of metrics attaining a metric threshold indicating a threshold value for determining whether to scale stateful service components. A scaling request can then be provided to the stateful service to initiate the scaling event at the stateful service in association with the set of components of the stateful service.

Abstract: A sign bit of a low-density parity-check (LDPC) codeword associated with a translation unit (TU) can be generated by performing an XOR operation on a RAIN drop corresponding to the TU and a raw read of the TU. The LDPC codeword can include a hard bit and three soft bits that include the sign bit. The LDPC codeword can be decoded using the hard bit and the three soft bits. A read recovery operation can be performed on the TU using the decoded LDPC codeword.

Abstract: Systems and methods are disclosed for advising a user when an energy storage device in a computing system needs charging. State of charge data of the energy storage device can be measured and stored at regular intervals. The historic state of charge data can be queried over a plurality of intervals and a state of charge curve generated that is representative of a user's charging habits over time. The state of charge curve can be used to generate a rate of charge histogram and an acceleration of charge histogram. These can be used to predict when a user will charge next, and whether the energy storage device will have an amount of energy below a predetermined threshold amount before the next predicted charging time. A first device can determine when a second device typically charges and whether the energy storage device in the second device will have an amount of energy below the predetermined threshold amount before the next predicted charge time for the second device.

Abstract: A computing environment can provide secure downloadable content regarding diagnostic test results to authorized and authenticated users. For example, a method disclosed herein can include receiving a diagnostic testing order for performing one or more tests. The method can also include generating downloadable content displaying test results. The method can further include generating a reference to a web resource including the downloadable content. Additionally, the method can include associating the reference with authorized users. The method can include delivering the reference for access by authorized users. The method can also include receiving a request for access to the web resource. Additionally, the method can include confirming the request is from an authenticated user. The method can further include confirming the request is from one of the authorized users. The method can include providing access to the web resource including the downloadable content to the authenticated and authorized user.

Abstract: The present disclosure describes methods of deacidifying and dechlorinating an aqueous solution using a deacidifying and dechlorinating composition. The present methods advantageously produce a dechlorinated brine solution have a neutral to alkaline pH.

Abstract: The present disclosure relates to utilizing a host failure recovery system to efficiently and accurately determine the health of host devices. For example, the host failure recovery system detects when a host server is failing by utilizing a power failure detection model that determines whether a host server is operating in a healthy power state or an unhealthy power state. In particular, the host failure recovery system utilizes a multi-layer power failure detection model that determines power-draw failure events on a host device. The failure detection model determines, with high confidence, the health of a host device based on power-draw signals and/or usage characteristics of the host device. Additionally, the host failure recovery system can initiate a quick recovery of a failing host device.

Abstract: A distributed function data transformation system includes a distributed function packet provisioning device coupled to compute systems. The distributed function packet provisioning device receives a request to perform a data transformation including data transformation operations, and determines a respective function for performing each data transformation operation. The distributed function packet provisioning device generates a distributed function packet including a function list identifying the respective functions, a data identifier identifying data upon which the respective functions should be performed, and a function performance identifier configured to identify one of the respective functions to perform on the data.

Abstract: A method and system for providing a virtual assistant to participants in a meeting. The method includes receiving a set of data associated with the meeting. The set of data includes meeting details, participant details, discussion details, audio details, and video details. Next, the method includes analyzing, using a trained model, the set of data to provide the virtual assistant to the participants in the meeting. Thereafter, the method includes assisting the participants in the meeting based on the analysis of the set of data.

Abstract: Methods, systems, and devices for data management are described. A data management system (DMS) may receive an indication of a storage configuration for an object storage location within a cloud environment. The DMS may generate a snapshot of a computing system in accordance with a backup policy associated with the computing system. The DMS may cause one or more data objects corresponding to the snapshot to be stored in the object storage location within the cloud environment in accordance with the storage configuration. The DMS may apply immutable object locks to the one or more data objects within the object storage location. The DMS may execute, in accordance with the backup policy associated with the computing system, one or more jobs that cause the respective immutable object locks for the one or more data objects to be extended one or more times during a retention period for the snapshot.

Abstract: Methods and systems for managing list data structures in an event-driven system are disclosed. One or more list data structures store notifications for a plurality of event consumers. One or more real-time event streams are monitored and, based on the monitoring, a first event is identified. One or more list entries are added to at least one list data structure of the one or more list data structures, to notify two or more event consumers of the plurality of event consumers of the first event. The one or more real-time event streams are further monitored, and, based on the monitoring, a further event associated with the first event is identified. Responsive to identifying the further event associated with the first event, the one or more list entries are updated to reflect the further event.

Abstract: A method of providing congestion control and reducing latency of data incoming to a core network, the method performed by a control plane gateway, includes: monitoring values of key performance indicators (KPIs) associated with a plurality of user plane gateways in the core network; predicting, using a machine learning (ML) model, an optimal window size respectively for each of the plurality of user plane gateways, based on the monitored values of the KPIs; and transmitting the optimal window size to the respective user plane gateway in the plurality of user plane gateways.

Abstract: In an example embodiment, a method of reducing SR max retransmissions in a communication network is disclosed. The method includes: accessing an SR max retransmission UE pool stored in a storage of the base station, the SR max retransmission UE pool being a list of plurality of UEs which experienced SR max retransmission. The method further includes identifying at least one UE in the SR max retransmission pool that would enter an SR max retransmission state, based on a last PUSCH uplink time period and a SR max retransmission time period associated with the at least one UE. Further, the method includes allocating a dummy UL grant to the identified at least one UE.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training a machine learning model using a deterministic data pipeline. One of the methods may include receiving a first request to generate a deterministic training dataset: transforming raw training examples obtained from the raw data source into pre-processed training examples; assigning a unique index to each pre-processed training example; and caching the pre-processed training examples into the cache directory specified in the received first request; receiving a second request to use the deterministic training dataset to train a machine learning model, the second request specifying a start index; and in response to receiving the second request: reading, from the cache directory, the pre-processed training examples that have indices beginning from the start index; and providing the read training examples in an order of the assigned indices for use in training the machine learning model.

Abstract: An antibacterial air freshening product for an interior environment. The product includes a container containing 1 ml to 50 ml of a freshening composition in fluid communication with a membrane. The composition is substantially free of a surfactant and has from 0.5% to 20% of a volatile aldehyde mixture, by weight of the composition. The volatile aldehyde mixture includes a C5 to C8 unbranched unsubstituted linear alkenal and a C9 to C14 unbranched unsubstituted linear alkenal. A weight ratio of the C5 to C8 unbranched unsubstituted linear alkenal to the C9 to C14 unbranched unsubstituted linear alkenal is from 3:1 to 1:3.

Abstract: An operation-security system for an automated vehicle includes an object-detector and a controller. The object-detector includes at least three sensors. Each sensor is one of a camera used to determine an image-location of an object proximate to a host-vehicle, a lidar-unit used to determine a lidar-location of the object proximate to the host-vehicle, and a radar-unit used to determine a radar-location of the object proximate to the host-vehicle. The controller is in communication with the at least three sensors. The controller is configured to determine a composite-location based on a comparison of locations indicated by the at least three sensors. Information from one sensor is ignored when a respective location indicated by the one sensor differs from the composite-location by greater than an error-threshold. If a remote sensor not on the host-vehicle is used, V2V or V2I communications may be used to communicate a location to the host-vehicle.