Abstract: Techniques and mechanisms for managing workloads in compute clusters comprising compute nodes by managing the workloads at the resource level of the compute clusters. For example, virtual service contexts (VSCs) may be defined where the VSCs represent service classes. Policies may be defined with respect to each service class. These service classes are dynamically constructed based on business needs. Hence there is natural requirement for a user to construct and rebalance the compute resources for these service classes dynamically. The policies may be related to resources of the compute clusters for executing workload units in the compute clusters. Resources of the compute clusters may be allocated to each service class. Each workload unit may be assigned to a one of the service classes based on the service context or type of workload unit. The workload units may then be executed by the compute clusters using the resources in accordance with the policies.

Abstract: A quick connect and disconnect system for a fluid coupling includes an axially extending fitting configured to be secured to a fluid conduit, a clip, and a quick connect and disconnect adaptor. The clip is coupled to the fitting and includes a base, a wing extending from the base, and a first lug extending from the first wing. The quick connect and disconnect adaptor includes a first release tab configured to resiliently deflect through an aperture of the adaptor. The adaptor is configured to transfer a compressive force to resiliently deflect the first lug to move the clip in an axial direction from an installed position to an uninstalled position.

Abstract: Existing approaches for switching between different hardware accelerators in a heterogeneous accelerator approach have the disadvantage that complete potential of the heterogeneous hardware accelerators do not get used as the switching relies on load on the accelerators or a random switching in which entire task gets reassigned to a different hardware accelerator. The disclosure herein generally relates to data model training, and, more particularly, to a method and system for data model training using heterogeneous hardware accelerators. In this approach, the system switches between hardware accelerators when a measured accuracy of the data model after any epoch is below a threshold of accuracy.

Abstract: Techniques described herein include frameworks and models for identifying, analyzing, and addressing hangs within distributed and heterogenous computing environments. A hang detection framework may model a distributed computing environment as a complex forest of interrelated requests. The hang detection framework may generate hang graphs based upon requests that are being processed and/or waited upon within the distributed environment. For example, a node within an acyclic graph may represent an execution entity that is currently processing one or more requests. Directed edges that connect one node to another may represent requests upon which an execution entity is waiting for another execution node to fulfill. The model may be used to isolate and address the root cause of hangs within the computing environment.

Abstract: Techniques and systems are provided for training a machine learning (ML) model. A technique can include generating a first set of features for objects in images, predicting image feature labels for the first set of features, comparing the predicted image feature labels to ground truth image feature labels to evaluate a first loss function, perform a perspective transform on the first set of features to generate a birds eye view (BEV) projected image features, combining the BEV projected image features and a first set of flattened features to generate combined image features, generating a segmented BEV map of the environment based on the combined image features, comparing the segmented BEV map to a ground truth segmented BEV map to evaluate a second loss function, and training the ML model for generation of segmented BEV maps based on the evaluated first loss function and the evaluated second loss function.

Abstract: The present invention relates to a process for refinery wastewater treatment. More particularly, the present invention relates to an automated process for treatment of refinery wastewater. The process of the present invention provides complete automation for controlling different critical parameters that enhance biological activity of activated sludge process (ASP) and helps in significant reduction in sludge recycling that increases the treatment efficiency.

Abstract: New data structures for analyzing a log are generated. A hierarchical data structure includes a plurality of hierarchical nodes. Each node is associated with data and metadata. Each node may also be associated with analysis data. Information (data, metadata, and/or analysis data) of an ancestor node is imputed to a descendant node; a descendant node inherits the information of an ancestor node. When determining analysis data for a particular hierarchical node, information from any ancestor node to the particular hierarchical node may be used; however, information from non-ancestor nodes is not necessarily used. A timeline data structure includes a reference to a hierarchical node within a hierarchical data structure and a reference to an event type. The timeline data structure is thereby associated with the information of the referenced hierarchical node and information of any ancestor nodes to the referenced hierarchical node.

Abstract: Systems, apparatuses, and methods for scanning a shopping space. The system pace comprises a locomotion system of a motorized unit, a camera system, and a central computer system. The central computer system is configured to receive motorized unit location information from the locomotion system, the motorized unit location information comprises a plurality of sets of coordinates each associated with a timestamp, receive item identifier instances from the camera system, wherein each item identifier instance is associated with a timestamp, and determine a location for each item identifier instance based on the timestamps associated of the item identifier instances and the timestamps associated with the plurality of sets of coordinates of the motorized unit location information.

Abstract: Various embodiments are directed to reconfigurable motorized conveyor rollers and methods of using the same. In various embodiments, a reconfigurable motorized conveyor roller comprises a housing assembly defining a roller width defined between a first lateral end and a second lateral end opposite the first lateral end; and a housing adjustment assembly disposed within the housing assembly that is operable to adjust the roller width by facilitating a configuration of the housing assembly between a nominal configuration and an expanded configuration; wherein the housing assembly comprises a plurality of coaxial housing components configured for movement relative to one another in one or more lateral directions, and wherein the housing adjustment assembly is operable to cause at least a portion of the plurality of housing components to move in the one or more lateral directions to selectively configure the housing assembly between the nominal configuration and the expanded configuration.

Abstract: Apparatuses, methods and systems comprising integrated motorized conveyor rollers are provided. The example integrated motorized conveyor roller comprises: a housing; a motor assembly and a drive assembly at least partially disposed within the housing that are configured to cause rotation of at least a portion of the integrated motorized conveyor roller; at least one sensing element operatively coupled to the integrated motorized conveyor roller; and a controller component in electronic communication with the at least one sensing element, the motor assembly and the drive assembly, wherein the controller component is configured to obtain operational data via the at least one sensing element.

Abstract: An optical master is created by using a nanoimprint alignment layer to pattern a liquid crystal layer. The nanoimprint alignment layer and the liquid crystal layer constitute the optical master. The optical master is positioned above a photo-alignment layer. The optical master is illuminated and light propagating through the nanoimprinted alignment layer and the liquid crystal layer is diffracted and subsequently strikes the photo-alignment layer. The incident diffracted light causes the pattern in the liquid crystal layer to be transferred to the photo-alignment layer. A second liquid crystal layer is deposited onto the patterned photo-alignment layer, which subsequently is used to align the molecules of the second liquid crystal layer. The second liquid crystal layer in the patterned photo-alignment layer may be utilized as a replica optical master or as a diffractive optical element for directing light in optical devices such as augmented reality display devices.

Abstract: Some embodiments provide retail inventory identification systems, comprising: a mobile product mapping system comprising: a transport system, a tower housing, and a cluster processing system comprising: a plurality of product identifying (PI) systems each in different orientations and directed along different fields of view, and a mobile control circuit; wherein the PI systems each comprise an image capture system, and a location determination system, wherein each of the plurality of PI systems is configured to: capture images of products supported on product support systems; and locally process the images comprising applying a plurality of different machine learning modeling techniques, to infer identification of multiple products and generate product identifying information in association with location information; and the mobile control circuit is configured to generate product identifying mapping of products as the product mapping system moves; and communicate the generated product identifying mapping to

Abstract: An optical device for characterizing flicker from an ambient light source is disclosed. The device comprises a radiation-sensitive device configured to vary a current in response to incident radiation, a second or higher-order modulator configured to output data corresponding to the current, and processing circuitry configured to provide a real-time characterization of flicker in the incident radiation based upon an analysis of changes in the data. Also disclosed is an associated method of characterizing flicker from an ambient light source.

Abstract: The present invention extends to methods, systems, and computer program products for deriving unified insights ad logs from DevOps Cl/CD tools and pipeline data. In general, a data transformer facilitates data normalization and serialization converting raw data across multiple DevOps tools and stores the data into a Data Lake in accordance with a customized schema. A continuous orchestrator sequences, aggregates and contextualizes the logs, providing an intuitive way of troubleshooting issues across a DevOps environment, historical data for compliance and audit purposes, and a build manifest for root cause analysis. The continuous orchestrator also processes the logs and leverages a KPI framework, providing intelligent dashboards across 90+ KPI's and a plurality of different dimensions (Planning, Development/pipelines, security, quality, operations, productivity and source code) to help customers make smart decisions and do more with less.

Abstract: Apparatuses, methods and systems comprising conveyor roller integrated divert mechanisms are provided. The example conveyor roller integrated divert mechanisms are integrated with a motorized or non-motorized conveyor roller to controllably divert an item at any point along a conveyor line. An example conveyor roller includes a housing forming an at least partially hollow cylindrical tube. The housing includes a plurality of radial gaps. In addition, the conveyor roller includes a divert assembly at least partially disposed within the housing that is operable to divert an item. The divert assembly includes a retractable divert mechanism configured to controllably protrude from the housing through at least one of the plurality of radial gaps.

Abstract: In one embodiment an optical sensing arrangement includes a first sensor configured to provide a first sensor signal, a second sensor configured to provide a second sensor signal, an integration unit including a first input which is connected to the first sensor, a second input which is connected to the second sensor, a first output which is configured to provide a first integration signal as a function of the first sensor signal, and a second output which is configured to provide a second integration signal as a function of the second sensor signal, a comparing unit including a first input which is connected to the first output of the integration unit, a second input which is connected to the second output of the integration unit and an output configured to provide a comparison signal as a function of the first and the second integration signal, and a control unit including a first input which is coupled to the output of the comparing unit, wherein the control unit is configured to evaluate pulses of the com

Abstract: This application relates to a system, method, and non-transitory computer readable medium for implementing a queuing platform. In some embodiments, a user request is received and it determined whether the user request matches predetermined criteria. Embodiments can include assigning the user request to an item queue and issuing a ticket. In further embodiments, the ticket is validated and an access token is transmitted to the user device based on the validation of the ticket. A checkout request corresponding to the at least one user request is received and validated. In some embodiments, the user request is transmitted from the item queue to a checkout engine to complete a purchase of the item based on the validation of the access token received in the checkout request.

Abstract: Some embodiments provide systems to identify products comprising: product vector database; a plurality of portable computing devices comprising a camera and a control circuit configured to: access an image captured by the camera; perform an optical character recognition on the image; apply a vector modeling rule to key text, generate a first query product vector and wirelessly communicate the first query product vector to the product recommendation system; the product recommendation system is configured to apply a vector evaluation rule to the first query product vector to identify a first product; and wirelessly communicate to the portable computing device the first product identifier; wherein the control circuit receives the first product identifier, accesses a product information, causes product information to be displayed; and causes the first product to be virtually added to a virtual cart.

Abstract: Apparatuses, methods, and systems comprising reconfigurable motorized conveyor rollers are provided. The example reconfigurable motorized conveyor roller includes a housing with a first end and a second end opposite to the first end. The housing includes a plurality of curved plates that at least partially form a cylindrical tube. The reconfigurable motorized conveyor roller includes a motor assembly and a drive assembly at least partially disposed within the housing that are configured to cause rotation of at least a portion of the reconfigurable motorized conveyor roller. And, the reconfigurable motorized conveyor roller includes a housing adjustment assembly disposed within the housing that is operable to modify one or more dimensions of the housing.