Abstract: A microservice management method responds to receiving a function request, comprising a request for a particular function, by collecting function metadata for the particular function and sending the function request and the function metadata to a cloud side orchestrator. The function metadata may include one or more prerequisite for an execution environment. Suitable microservices for the particular function are identified, in accordance with the function metadata. The identified microservices are sent to an edge side orchestrator that installs, initializes, and monitors these microservices and, responsive to detecting complete execution of the particular function, terminates one or more of the suitable microservices.

Abstract: Disclosed subject matter includes a testing method that accesses tracing data, comprising a plurality of tracing records corresponding to a plurality of test cases executed on a target system. The target system includes a plurality of microservices configured in accordance with a REST API. A dynamic test analyzer accesses the tracing data and imports a definition of the REST API. The analyzer checks each tracing record for consistency with the API definition. Based on this checking of the tracing records, one or more coverage metrics may be calculated. If any one or more of the coverage metrics fails to exceed a threshold coverage, a testing action may be required. In some cases, the testing action may be selected from either seeking additional coverage of the APIs included in the target system or seeking additional test result coverage.

Abstract: A disclosed microservice scaling operation obtains information indicating dependencies between a function associated with an external API call and microservices spanned by the external API call. Functions invoked by managed resources are monitored and, responsive to detecting the function being invoked, a scaling service is launched to access the dependency information, identify the applicable microservices, and perform a scale up operation instantiating each of the microservices. The dependency information may be obtained by recording and analyzing traces for instances of the external API call to determine a dependency tree that indicates branches spanned by the external API call and a sequence of microservices corresponding to each branch. The microservices may be scaled up in parallel or in a prioritized parallel manner wherein in early span microservices are launched before late span microservices. The API may be a RESTful API and each microservice may correspond to an internal API call.

Abstract: A disclosed security broker receives a request message addressed to a message queue associated with an edge resource, identifies the message source, and leverages an authentication module to verify that the source has authorization to access the targeted message queue. The security broker may then deliver the request message to the targeted edge resource message queue. If the edge resource and security broker are physically isolated, the security broker may format the request message as a file, store the file to a storage device, and transfer the storage device to the edge resource, which may then process the file and upload the request message to the edge resource's message queue module. The security broker also monitors a response message from the edge resource, purges sensitive data from the response.

Abstract: A disclosed method for implementing containers in an information handling system generates, with a first non-privileged container, a request that is sent to a RESTful API. Whenever the API identifies a request requiring host access the API launches a second container, which is configured to operate in a privileged execution mode. The second container accesses the host and executes the requested actions. When the request completes, the first container resumes non-privileged execution, thereby confining privileged mode execution to a container that is only active during host interaction. The host access can be access required to: exchange data with the host, query the host for hardware information, and modify host configuration. The host may be implanted within a device featuring an HCI infrastructure. In one configuration, the host resides of one of multiple distinct nodes of an HCI appliance.

Abstract: An information handling system may include at least one processor and a memory. The information handling system may be configured to: execute an application comprising a plurality of microservices on the at least one processor, wherein the application is configured to service external requests by executing a plurality of application programming interface (API) calls among the plurality of microservices; and for each API call: determine a span ID associated with such API call; determine a trace ID associated with a particular external request that is associated with such API call; and log the span ID and the trace ID.

Abstract: The application discloses a power supply device, including: a circuit board having a first side and a second side opposite to each other, the circuit board at least mounted with a first heating element disposed close to the second side; a fan disposed on the first side of the circuit board for providing an air-cooled airflow for heat dissipation, wherein an airflow direction is defined from the first side to the second side; a first airflow guide member disposed in parallel to the first heating element; and a second airflow guide member at least partially disposed above the first heating element, and adjacent to the first airflow guide member, wherein the first airflow guide member, the second airflow guide member and the first heating element form a first auxiliary air channel; wherein a direction of the first auxiliary air channel is the same as the airflow direction.

Abstract: An experimental apparatus comprises a centrifuge basket experiment module, a structural deformation device, an image reflector and a geomorphometer; the centrifuge basket experiment module is installed on a cantilever of a drum centrifuge in a hanging manner during an experiment; the structural deformation device is configured for extending and compressing experimental materials; a mirror panel of the image reflector reflects images inside an experiment box through a transparent window; the geomorphometer can simulate rainfall and recover wastewater, and the apparatus can also heat the experimental materials. An image acquisition device is installed on a bulkhead of the experiment module and the cantilever of the centrifuge to acquire top and side images of the experiment box.

Abstract: A method and an apparatus for image capturing, and a storage medium can be applied to a terminal device. When it is determined that the terminal is in an augmented reality image capturing scene, augmented reality environment information of the augmented reality image capturing scene is determined. When an image capturing instruction is received, an image is captured and augmented reality environment information corresponding to the image is recorded to obtain a target image carrying the augmented reality environment information.

Abstract: The present disclosure discloses a deep spatial-temporal similarity method for air quality prediction, and belongs to the technical field of environmental protection. When the method predicts air quality-related indexes of a target site, a temporal change of air pollution and a spatial diffusion relationship are effectively combined, and then spatial-temporal similarity sites of the target site are selected; air quality monitoring data collected by the target site, the spatial-temporal similarity site of the target site and geographical neighbour sites of the target site and meteorological data are respectively taken as inputs of a long short term memory network (LSTM) model to obtain uncorrelated output results, and then predicted values of air quality-related index data of the target site are obtained in a mode of support vector regression (SVR) integration.

Abstract: A method for managing a cloud computing platform including an edge-side and a cloud-side sends a data collection service from the cloud side to the edge side environment to collect edge data indicative of one or more microservice architecture (MSA) parameters associated with one or more microservices in the edge side. An measurement of the edge data is performed to identify dependencies between one or more MSA parameters and one or more edge environment settings. Based at least in part on the MSA measurement, one or more edge environment optimizations are identified and one or more appropriate fixes are implemented within a cloud-side modified edge environment. Before distributing the modified edge environment to the edge, an in-cloud verification of the cloud-side modified edge environment.

Abstract: A team photographing method can be applied to terminals. The terminals can include a main terminal and at least one member terminal. The method includes: forming a team including the main terminal and the at least one member terminal based on a team photographing instruction initiated by the main terminal; based on a target object to be photographed determined by the main terminal, determining the target object to be photographed of the at least one member terminal; and controlling, by the main terminal, the main terminal and the at least one member terminal to photograph the target object to be photographed cooperatively.

Abstract: The present disclosure discloses a deep spatial-temporal similarity method for air quality prediction, and belongs to the technical field of environmental protection. When the method predicts air quality-related indexes of a target site, a temporal change of air pollution and a spatial diffusion relationship are effectively combined, and then spatial-temporal similarity sites of the target site are selected; air quality monitoring data collected by the target site, the spatial-temporal similarity site of the target site and geographical neighbour sites of the target site and meteorological data are respectively taken as inputs of a long short term memory network (LSTM) model to obtain uncorrelated output results, and then predicted values of air quality-related index data of the target site are obtained in a mode of support vector regression (SVR) integration.

Abstract: A display device is provided, including: a display panel; a backlight module; a circuit board, arranged on a side of the backlight module facing away from the display panel; a flexible circuit board, connected to the circuit board and the display panel, where the flexible circuit board includes a bending portion arranged on a side of the backlight module and extending from the display panel to the circuit board; a circuit board protective frame, arranged around the circuit board and the flexible circuit board, where an opening is provided on a first portion of the circuit board protective frame, and the first portion is a portion of the circuit board protective frame facing the bending portion; and a buffer structure, fixed on the circuit board protective frame, arranged parallel to the circuit board protective frame, and covering the opening.

Abstract: Disclosed is a real-time object detection method deployed on a platform with limited computing resources, which belongs to the field of deep learning and image processing. In the present invention, YOLO-v3-tiny neural network is improved, Tinier-YOLO reserves the front five convolutional layers and pooling layers of YOLO-v3-tiny and makes prediction at two different scales. Fire modules in SqueezeNet, 1×1 bottleneck layers, and dense connection are introduced, so that the structure is used to achieve smaller, faster, and more lightweight network that can be run in real time on an embedded AI platform. The model size of Tinier-YOLO in the present invention is only 7.9 MB, which is only ¼ of 34.9 MB of YOLO-v3-tiny, and ? of YOLO-v2-tiny. The reduction in the model size of Tinier-YOLO does not affect real-time performance and accuracy of Tinier-YOLO. Real-time performance of Tinier-YOLO in the present invention is 21.8% higher than that of YOLO-v3-tiny and 70.8% higher than that of YOLO-v2-tiny.

Abstract: A quantization apparatus includes: an imaging unit that converts statistical data at a position in a real space into a pixel value of a coordinate, which is associated with the position, in an image; and a derivation unit that derives a quantization parameter corresponding to a quantization width of the pixel value for each of one or more positions in the real space with respect to a part or a whole of the image. The quantization apparatus may further include a purpose acquisition unit that acquires information indicating a purpose related to statistics. The derivation unit may derive the quantization parameter corresponding to the quantization width that is larger as the pixel value is larger.

Abstract: The present disclosure relates to a chip on film and a display device. The chip on film includes a body and an insulating protective film arranged on the body, in which the body includes a first area, a first binding area for binding and connecting to the back surface of the display panel, and a first bendable area located between the first area and the first binding area and capable of being bent in a first direction; and the insulating protection film includes a first connection area connected to the first area, a second connection area for connecting to the back surface of the display panel, and a second bendable area located between the first connection area and the second connection area and capable of being bent in a second direction opposite to the first direction.

Abstract: A color filter substrate, a manufacturing method thereof, and related devices. The color filter substrate includes: a base plate; a color filter layer on the base plate; and a transparent layer on a side of the color filter layer away from the base plate. Specifically, the color filter layer includes multiple rows of color resist units, wherein color resist units in each row have at least three different colors. In addition, a first distance between a surface of the base plate close to the color resist units and a surface of the transparent layer away from the color resist units may be different for color resist units of different colors.

Abstract: A capturing method for an electronic device, includes: obtaining an image captured during a capturing process, obtaining a target pop-up comment matching image information of the image, and displaying the image and the target pop-up comment.

Abstract: A team photographing method can be applied to terminals. The terminals can include a main terminal and at least one member terminal. The method includes: forming a team including the main terminal and the at least one member terminal based on a team photographing instruction initiated by the main terminal; based on a target object to be photographed determined by the main terminal, determining the target object to be photographed of the at least one member terminal; and controlling, by the main terminal, the main terminal and the at least one member terminal to photograph the target object to be photographed cooperatively.