Abstract: Examples described herein include one or more processors; a network interface; and a direct memory access (DMA) engine communicatively coupled to the one or more processors. In some examples, the DMA engine is to receive a DMA data access request and based on an address in the DMA data access request corresponding to a remote memory device, the DMA engine is to cause the network interface to generate at least one packet for transmission to the remote memory device. In some examples, if the source address corresponds to a local memory device and the destination address corresponds to a remote memory device, the DMA engine is to cause the network interface to generate at least one packet for transmission to the remote memory device.

Abstract: Examples described herein relate to memory thin provisioning in a memory pool of one or more dual in-line memory modules or memory devices. At any instance, any central processing unit (CPU) can request and receive a full virtual allocation of memory in an amount that exceeds the physical memory attached to the CPU (near memory). A remote pool of additional memory can be dynamically utilized to fill the gap between allocated memory and near memory. This remote pool is shared between multiple CPUs, with dynamic assignment and address re-mapping provided for the remote pool. To improve performance, the near memory can be operated as a cache of the pool memory. Inclusive or exclusive content storage configurations can be applied. An inclusive cache configuration can include an entry in a near memory cache also being stored in a memory pool whereas an exclusive cache configuration can provide an entry in either a near memory cache or in a memory pool but not both.

Abstract: A system comprising a traffic handler comprising circuitry to determine that data of a memory request is stored remotely in a memory pool; generate a packet based on the memory request; and direct the packet to a path providing a guaranteed latency for completion of the memory request.

Abstract: A method of identifying sector load imbalance. The method includes identifying a first set of imbalanced cells in a sector based on a first condition, wherein the first condition is based on network usage of the cells in the sector. The method further includes identifying a second set of imbalanced cells from the first set of imbalanced cells based on a second condition, wherein the second condition is based on one or more physical parameters associated with each of the first set of imbalanced cells. The method further includes displaying visualizations associated with the second set of imbalanced cells, the visualizations indicating miscalculations of the one or more physical parameters.

Abstract: Examples described herein and includes at least one processor and a direct memory access (DMA) device. In some examples, the DMA device is to: access a command from a memory region allocated to receive commands for execution by the DMA device, wherein the command is to access content from a local memory device or remote memory node. In some examples, the DMA device is to: determine if the content is stored in a local memory device or a remote memory node based on a configuration that indicates whether a source address refers to a memory address associated with the local memory device or the remote memory node and whether a destination address refers to a memory address associated with the local memory device or the remote memory node. In some examples, the DMA device is to: copy the content from a local memory device or copy the content to the local memory device using a memory interface.

Abstract: Systems and methods for describing a composition of an article of manufacture are disclosed. In one aspect, a method includes receiving article composition data for an article of manufacture that identifies a set of parts of the article, a stated composition for each part of the set of parts, and a physical quantity of the stated composition. The method further includes classifying the stated composition of each part of the set of parts into a normalized composition that includes a set of normalized chemicals. The method further includes outputting an aggregated physical quantity of each normalized chemical for the set of parts of the article. The method can include classifying a normalized composition of each part into a material category within a hierarchical taxonomy based on the set of normalized chemicals of that normalized composition and outputting an aggregated physical quantity of each material category for the parts.

Abstract: A device may receive user activity data identifying activities of a user, and content data identifying text transcripts associated with the user. The device may generate a first custom embedding associated with the user based on the user activity data, and may process the first custom embedding, with a machine learning model, to generate an intent of the user and a next action for the user based on the intent. The device may process the content data to generate a second custom embedding for the user and an end user vector based on the second custom embedding, and may generate a document vector for the user based on the next action for the user. The device may process the document vector and the end user vector, with a language model, to generate a document for the user, and may perform one or more actions based on the document.

Abstract: A cache cleaner controller is described. In one or more examples, an apparatus includes a cache directory including status bits associated with cache locations within cache storage and a cache cleaner controller. The cache cleaner controller is configured to detect that a cache cleaner threshold has been reached. The cache cleaner threshold defines that a threshold number of the status bits indicate data maintained at the cache locations, respectively, has been changed. The cache cleaner controller is also configured to cause the data indicated as changed by the status bits to be copied from the cache locations within cache storage to the physical volatile memory.

Abstract: Embodiments herein provide a method for handling redeployment of an existing application build. The method includes receiving a build creation command message for redeployment of the existing application build. Further, the method includes determining availability of a module input, a client input, a workspace input, and an application input in the build creation command message. Further, the method includes determining specific chunks of the existing application build for redeployment based on the module input, the client input, the workspace input, and the application input available in the build creation command message. Further, the method includes redeploying the specific chunks of the existing application build.

Abstract: An illustrative intent classification engine may access a text transcript and determine one or more features associated with the text transcript. Based on the one or more features, the intent classification engine may generate an aggregate embedding vector and provide the aggregate embedding vector as an input to a trained model configured to output an intent classification. Corresponding methods and systems are also disclosed.

Abstract: Systems and methods for describing a composition of an article of manufacture are disclosed. In one aspect, a method includes receiving article composition data for an article of manufacture that identifies a set of parts of the article, a stated composition for each part of the set of parts, and a physical quantity of the stated composition. The method further includes classifying the stated composition of each part of the set of parts into a normalized composition that includes a set of normalized chemicals. The method further includes outputting an aggregated physical quantity of each normalized chemical for the set of parts of the article. The method can include classifying a normalized composition of each part into a material category within a hierarchical taxonomy based on the set of normalized chemicals of that normalized composition and outputting an aggregated physical quantity of each material category for the parts.

Abstract: The present teaching relates to customer service with AI-based automated auditing on agent fraud. Real-time features of a communication between an agent and a customer are obtained. To detect agent fraud, a batch feature vector is computed based on real-time features extracted from communications involving the agent and accumulated over a batch period. Agent fraud is detected based on a model and the detection result is used to audit the agent for service performance.

Abstract: A high strength and biodegradable polymer film comprising a blend of a polylactic acid (PLA) copolymer, a flexible polymer linker, and optionally a compostable polyester segment is disclosed. Upon biaxially stretching, the disclosed polymer film exhibits unexpectedly high tensile strengths and high impact strengths while providing high flexibility as measured according to various standards such as ASTM D882, D3420, and D1709. The disclosed films are formulated to meet common industrial composting standards as defined by ASTM D6400, EN 13432, and ISO 17088, as well as exhibit accelerated biodegradation rates at lower temperatures for home composting applications.

Abstract: Provided are a system and method of decongesting a highly utilized cell within a sector of a wireless telecommunications network. The method includes receiving a set of reference samples taken from a plurality of points within the wireless telecommunications network within a predetermined time period, calculating, based on the set of reference samples, a first data volume comprising a sum of a total data volume for all highly utilized cells within the sector containing the highly utilized cell, calculating, based on the set of reference samples, a first total expected traffic offload value, determining whether the first total expected traffic offload value is greater than a predetermined percentage of the first data volume, and based on determining that the first total expected traffic offload value is greater than the predetermined percentage of the first data volume, generating an output for decongesting the cell.

Abstract: A method includes receiving a coverage hole mitigation work order (CHMWO) created by a project work order (PWO) administrator; receiving assignment of the CHMWO from the PWO administrator to a PWO requester that is included in a geographic location of the CHMWO; receiving coverage hole optimization for the CHMWO from the PWO requester; and closing the CHMWO in response to the coverage hole optimization being validated.

Abstract: A method for service gap mitigation via tilt recommendation may include: identifying a candidate cell; generating boundaries of a service gap; determining a best server plot of the candidate cell; determining geo-located samples of the candidate cell; categorizing the candidate cell based on the boundaries of the service gap and at least one of the best server plot or the geo-located samples; and calculating an electronic tilt based on the categorization.

Abstract: Methods and systems for optimizing antenna tilt in a cell site of a radio access network (RAN), the system includes: a memory storing instructions; and a processor configured to execute the instructions to: obtain input data for a source cell and for neighboring cells thereof at a plurality of predetermined geolocations; determine an overlap percentage from the input data obtained at the plurality of predetermined geolocations; determine, for the source cell, at least one TA percentage for at least one distance range from a center of the source cell; determine, based on the determined overlap percentage and the determined at least one TA percentage, that the source cell is outside a predetermined cell coverage area; and recommend, based on the determining that the source cell is outside the predetermined cell coverage area, adjustment of the antenna tilt for the source cell to meet the predetermined cell coverage area.

Abstract: Embodiments herein provide a method for prioritizing booming cells in a wireless network by an electronic device. The method includes receiving information about a plurality of booming cells. Further, the method includes determining a priority for each booming cell of the plurality of booming cells based on a morphology factor, a PRB utilization factor, a RRC connected users factor, a booming distance, and a user density factor. Further, the method includes prioritizing the plurality of booming cells for mitigation based on the priority for each booming cell of the plurality of booming cells.

Abstract: A method of identifying call muting areas includes dividing a network cluster into a plurality of grids, determining a call muting percentage of at least one first grid of the plurality of grids, determining whether the call muting percentage of the at least one first grid is greater than a predetermined call muting percentage threshold, and generating at least one call muting polygon including at least a portion of the at least one first grid based on determining that the call muting percentage of the at least one first grid is greater than the predetermined call muting percentage threshold

Abstract: Embodiments herein provide a method for handling redeployment of an existing application build. The method includes receiving a build creation command message for redeployment of the existing application build. Further, the method includes determining availability of a module input, a client input, a workspace input, and an application input in the build creation command message. Further, the method includes determining specific chunks of the existing application build for redeployment based on the module input, the client input, the workspace input, and the application input available in the build creation command message. Further, the method includes redeploying the specific chunks of the existing application build.