Abstract: The present disclosure relates to a method, a device, and a program product for managing a computing system. In a method, a current state and a plurality of historical states of a computing device in a computing system are acquired, the plurality of historical states respectively describing historical states of the computing device in the computing system at a plurality of historical time points. In response to determining that the current state matches a scheduling type for scheduling the computing device, the plurality of historical states are searched for a historical state matching the current state. A historical scheduling policy associated with the historical state is determined. Based on the historical scheduling policy, a computing task to be executed by the computing device is allocated to at least one other computing device in the computing system.

Abstract: A graphics processing unit (GPU) implements operations, with associated op codes, to perform mixed precision mathematical operations. The GPU includes an arithmetic logic unit (ALU) with different execution paths, wherein each execution path executes a different mixed precision operation. By implementing mixed precision operations at the ALU in response to designate op codes that delineate the operations, the GPU efficiently increases the precision of specified mathematical operations while reducing execution overhead.

Abstract: A processing unit such as a graphics processing unit (GPU) includes a plurality of vector signal processors (VSPs) that include multiply/accumulate elements. The processing unit also includes a plurality of registers associated with the plurality of VSPs. First portions of first and second matrices are fetched into the plurality of registers prior to a first round that includes a plurality of iterations. The multiply/accumulate elements perform matrix multiplication and accumulation on different combinations of subsets of the first portions of the first and second matrices in the plurality of iterations prior to fetching second portions of the first and second matrices into the plurality of registers for a second round. The accumulated results of multiplying the first portions of the first and second matrices are written into an output buffer in response to completing the plurality of iterations.

Abstract: An apparatus comprises at least one processing device configured to establish, utilizing a communication network, a remote collaboration session between first and second computing devices associated with first and second users, and to determine data to be exchanged between the first and second computing devices as part of the remote collaboration session. The first and second computing devices each comprise at least one of an augmented reality and a virtual reality computing device. The processing device is also configured to generate different logical network slices of the communication network, each logical network slice having a set of network characteristics associated with at least one type of the data to be exchanged between the first and second computing devices as part of the remote collaboration session. The processing device is further configured to communicate the different types of data between the first and second computing devices utilizing the logical network slices.

Abstract: A transformer includes a primary winding, a feedback winding, and a non-feedback winding having substantially equal winding widths in an axial direction of a bobbin.

Abstract: A parallel processing unit employs an arithmetic logic unit (ALU) having a relatively small footprint, thereby reducing the overall power consumption and circuit area of the processing unit. To support the smaller footprint, the ALU includes multiple stages to execute operations corresponding to a received instruction. The ALU executes at least one operation at a precision indicated by the received instruction, and then reduces the resulting data of the at least one operation to a smaller size before providing the results to another stage of the ALU to continue execution of the instruction.

Abstract: Embodiments of the present disclosure relate to a method, an electronic device, and a computer program product for processing a query request. The method includes determining a first number of requests of a first edge device and a second number of requests of a second edge device in a distributed system. The method further includes determining the first edge device as a parent node and the second edge device as a child node in response to both the first number of requests and the second number of requests being less than a threshold. The method further includes sending to the child node an indication of transmitting data in the child node to the parent node. The method further includes sending, in response to receiving a query request for the child node, the query request to the parent node.

Abstract: This application discloses an access management method, an authenticator, and an authentication server, applied to a scenario in which a terminal device accesses a network. After completing authentication, a terminal device sends a first packet to an authenticator, where the first packet carries a first IPv6 address of the terminal device and a MAC address of the terminal device. When determining that the first IPv6 address is a new IPv6 address, the authenticator sends, to an authentication server, a second packet carrying the first IPv6 address and the MAC address, so as to indicate the authentication server to send a first authorization policy to a policy enforcement point based on the first IPv6 address.

Abstract: The present invention provides a two-component polyurethane composition, and belongs to the technical field of high polymer material. The two-component polyurethane composition comprises component A and component B. According to mass percent, the component A comprises 0-100% of high activity polyether polyol, 0-100% of cross-linking agent, 0-40% of chain extender, 0-50% of plasticizer and 0-5% of catalyst; according to mass percent, the component B comprises 0-100% of diphenylmethane diisocyanate and derivative thereof, and 0-50% of plasticizer; and the mass ratio of the component A to the component B is 1:0.1-1:3. The two-component polyurethane composition provided by the present invention has low heat release and excellent mechanical properties.

Abstract: Embodiments of the present disclosure relate to a method, an electronic device, and a computer program product for adjusting a computing load. The method in an illustrative embodiment includes: determining a total computing power demand of at least one user device that will be switched, due to movement, to being provided a computing service by a computing node; determining an available computing power of the computing node; and if the available computing power is unable to meet the total computing power demand, by adjusting a computing load of the computing node, adjusting the available computing power before the at least one user device is switched to being provided the computing service by the computing node, so as to meet the total computing power demand.

Abstract: A processing system executes wavefronts at multiple arithmetic logic unit (ALU) pipelines of a single instruction multiple data (SIMD) unit in a single execution cycle. The ALU pipelines each include a number of ALUs that execute instructions on wavefront operands that are collected from vector general process register (VGPR) banks at a cache and output results of the instructions executed on the wavefronts at a buffer. By storing wavefronts supplied by the VGPR banks at the cache, a greater number of wavefronts can be made available to the SIMD unit without increasing the VGPR bandwidth, enabling multiple ALU pipelines to execute instructions during a single execution cycle.

Abstract: A method for planting submerged plants is disclosed. The method includes the steps of selecting a combination of two submerged plants; planting the combination of submerged plants; measuring a biomass and evaluating the biomass. By discussing the influence of planting proportion on the biomass of submerged plant community, the structure of submerged plant community can be optimized, and the success rate of submerged vegetation restoration and reconstruction can be improved by using the positive interaction between plant species.

Abstract: The present disclosure provides a method for storing data, including: storing data in a first node, the first node being selected from a first node set including a first edge node and child nodes thereof; selecting, from a second node set including a second edge node and child nodes thereof, a second node for storing the data, the first edge node being a child node of the second edge node; and storing the data in the second node. The present disclosure further provides a method for searching for data. The present disclosure can reduce the time for searching for data in a distributed storage system.

Abstract: Implementations of the present disclosure relate to a method, an electronic device, and a computer program product for managing an inference process. Here, the inference process is implemented based on a machine learning model. A method includes: determining, based on a computational graph defining the machine learning model, dependency relationships between a set of functions for implementing the inference process; acquiring, in at least one edge device located in an edge computing network, a set of computing units available to execute the inference process; selecting at least one computing unit for executing the set of functions from the set of computing units; and causing the at least one computing unit to execute the set of functions based on the dependency relationships. With example implementations of the present disclosure, the inference process is implemented by making use of a variety of computing units in the edge computing network, thereby improving performance.

Abstract: Embodiments of the present disclosure provide a method for distributing virtual visual content, including: sending a first content portion in virtual visual content to be interacted in user equipment to a plurality of edge devices; selecting at least one edge device from the plurality of edge devices; and sending a second content portion in the virtual visual content to the selected at least one edge device, the second content portion having a higher change rate than the first content portion in the interaction. According to the embodiments of the present disclosure, a portion of virtual visual content can be distributed to selected edge devices in advance, and there is no need to distribute the virtual visual content to all edge devices, thereby saving network burden, reducing bandwidth requirements, and improving distribution efficiency.

Abstract: A processor includes a plurality of vector sub-processors (VSPs) and a plurality of memory banks dedicated to respective VSPs. A first memory bank corresponding to a first VSP includes a first plurality of high vector general purpose register (VGPR) banks and a first plurality of low VGPR banks corresponding to the first plurality of high VGPR banks. The first memory bank further includes a plurality of operand gathering components that store operands from respective high VGPR banks and low VGPR banks. The operand gathering components are assigned to individual threads while the threads are executed by the first VSP.

Abstract: A display substrate includes a substrate and a display region, an isolation region, a peripheral region and a hole disposed on the substrate, wherein the display region surrounds the hole, the isolation region is disposed between the display region and the peripheral region, and the peripheral region surrounds and adjoins the hole; the isolation region is provided with an isolation column and a first inorganic structure, the isolation column is disposed at a side of the substrate, and the first inorganic structure is disposed at a side of the isolation column away from the substrate; and the peripheral region is provided with a structure layer, and the structure layer is disposed on the substrate and includes a first organic structure and a second inorganic structure disposed along a direction perpendicular to the substrate; wherein the first inorganic structure, the second inorganic structure and the first organic structure all are made of an insulation material, and the first inorganic structure and the sec

Abstract: Methods and systems are disclosed for executing operations on single-instruction-multiple-data (SIMD) units. Techniques disclosed perform a dot product operation on input data during one computer cycle, including convolving the input data, generating intermediate data, and applying one or more transitional operations to the intermediate data to generate output data. Aspects described, wherein the input data is an input to a layer of a convolutional neural network and the generated output data is the output of the layer.

Abstract: The present disclosure provides a method and a system for modelling a poor texture tunnel based on a vision-lidar coupling. The method includes: obtaining point cloud information collected by a depth camera, laser information collected by a lidar, and motion information of an unmanned aerial vehicle (UAV); generating a raster map based on the laser information, and obtaining pose information of the UAV based on the motion information; obtaining a map model through fusing the point cloud information, the raster map, and the pose information by a Bayesian fusion method; and correcting a latest map model by feature matching based on a previous map model.

Abstract: Techniques are disclosed for generating a virtual representation (e.g., one or more digital twin models) of a multi-access edge computing system environment, and managing the multi-access edge computing system environment via the virtual representation. By way of example only, such techniques enable understanding, prediction and/or optimization of performance of applications and/or systems operating in the multi-access edge computing environment.