Abstract: The disclosure relates to a method, apparatus, system, medium and electronic device for graph neural network generation. The method includes: obtaining a subgraph structure, the subgraph structure being configured to reflect a graph structure of a corresponding subgraph, and the subgraph comprising a plurality of nodes and edges between the plurality of nodes; obtaining, based on the subgraph structure and according to a predetermined priority, node features of the plurality of nodes and edge features of the edges from a plurality of memories; the predetermined priority being obtained by sorting the plurality of memories in accordance with memory size in an ascending order; fusing, based on the subgraph structure, the node features of the plurality of nodes and the edge features of the edges to obtain subgraph data; and training, based on the subgraph data, the graph neural network.

Abstract: Embodiments of the disclosure provide a method and apparatus for running a service, and an electronic device. An embodiment of the method includes: determining a target deployment manner of a graphics processing unit (GPU) according to performance data of each service in a service set, where the deployment manner includes: dividing the GPU into sub-GPUs of a respective size and determining a service configured to be run by each sub-GPU; and switching, for the service in the service set, running of the service from a sub-GPU indicated by a current deployment manner to a sub-GPU indicated by the target deployment manner. According to the embodiment, waste of the GPU can be reduced by running a plurality of services on the GPU.

Abstract: A microdevice for monitoring a target analyte is provided. The microdevice can include a field effect transistor comprising a substrate, a gate electrode, and a microfluidic channel including graphene. The microfluidic channel can be formed between drain electrodes and source electrodes on the substrate. The microdevice can also include at least one aptamer functionalized on a surface of the graphene. The at least one aptamer can be adapted for binding to the target analyte. Binding of the target analyte to the at least one aptamer can alter the conductance of the graphene.

Abstract: Embodiments of the disclosure provide a method and apparatus for running a service, and an electronic device. An embodiment of the method includes: determining a target deployment manner of a graphics processing unit (GPU) according to performance data of each service in a service set, where the deployment manner includes: dividing the GPU into sub-GPUs of a respective size and determining a service configured to be run by each sub-GPU; and switching, for the service in the service set, running of the service from a sub-GPU indicated by a current deployment manner to a sub-GPU indicated by the target deployment manner. According to the embodiment, waste of the GPU can be reduced by running a plurality of services on the GPU.

Abstract: Distributed computing systems, devices, and associated methods of virtual RDMA switching are disclosed herein. In one embodiment, a method includes intercepting a command from an application in a container to establish an RDMA connection with a remote container on a virtual network. In response to the intercepted command, an RDMA endpoint at a physical NIC of a server is created. The method can also include intercepting another command to pair with a remote RDMA endpoint corresponding to the remote container. The intercepted another command contains data representing a routable network address of the remote RDMA endpoint in the RDMA computer network. Then, the RDMA endpoint created at the physical NIC of the server can be paired with the remote RDMA endpoint using the routable network address of the remote RDMA endpoint.

Abstract: A high k passivation layer, an anti-reflective coating layer, and a buffer layer are disposed over semiconductor substrate including photodiodes formed therein. Trenches are etched into the semiconductor substrate through the buffer layer, anti-reflective coating layer, and the high k passivation layer in a grid-like pattern surrounding each of the photodiodes in the semiconductor substrate. Another high k passivation layer lines an interior of the trenches in the semiconductor substrate. An adhesive and barrier layer is deposited over the high k passivation layer that lines the interior of the trenches. A deep trench isolation (DTI) structure is formed with conductive material deposited into the trenches over the adhesive and barrier layer to fill the trenches. A grid structure is formed over the DTI structure and above a plane of the buffer layer. The grid structure is formed with the conductive material.

Abstract: Proxy-based scaling may be performed for databases. A proxy may be implemented for a database that can establish a connection between the proxy and a database engine to perform a database queries received from a client at the proxy. A scaling event may be detected for the database responsive to which the proxy may establish a connection with a new database engine which may, in some embodiments, have different capabilities or resources that address the features or criteria that triggered the scaling event. Session state may be copied from the database engine to the new database engine so that the new database engine may be able to provide access to the database on behalf of requests received from the client through the proxy.

Abstract: A high k passivation layer, an anti-reflective coating layer, and a buffer layer are disposed over semiconductor substrate including photodiodes formed therein. Trenches are etched into the semiconductor substrate through the buffer layer, anti-reflective coating layer, and the high k passivation layer in a grid-like pattern surrounding each of the photodiodes in the semiconductor substrate. Another high k passivation layer lines an interior of the trenches in the semiconductor substrate. An adhesive and barrier layer is deposited over the high k passivation layer that lines the interior of the trenches. A deep trench isolation (DTI) structure is formed with conductive material deposited into the trenches over the adhesive and barrier layer to fill the trenches. A grid structure is formed over the DTI structure and above a plane of the buffer layer. The grid structure is formed with the conductive material.

Abstract: Distributed computing systems, devices, and associated methods of virtual RDMA switching are disclosed herein. In one embodiment, a method includes intercepting a command from an application in a container to establish an RDMA connection with a remote container on a virtual network. In response to the intercepted command, an RDMA endpoint at a physical NIC of a server is created. The method can also include intercepting another command to pair with a remote RDMA endpoint corresponding to the remote container. The intercepted another command contains data representing a routable network address of the remote RDMA endpoint in the RDMA computer network. Then, the RDMA endpoint created at the physical NIC of the server can be paired with the remote RDMA endpoint using the routable network address of the remote RDMA endpoint.

Abstract: Proxy-based scaling may be performed for databases. A proxy may be implemented for a database that can establish a connection between the proxy and a database engine to perform a database queries received from a client at the proxy. A scaling event may be detected for the database responsive to which the proxy may establish a connection with a new database engine which may, in some embodiments, have different capabilities or resources that address the features or criteria that triggered the scaling event. Session state may be copied from the database engine to the new database engine so that the new database engine may be able to provide access to the database on behalf of requests received from the client through the proxy.

Abstract: Distributed computing systems, devices, and associated methods of virtual RDMA switching are disclosed herein. In one embodiment, a method includes intercepting a command from an application in a container to establish an RDMA connection with a remote container on a virtual network. In response to the intercepted command, an RDMA endpoint at a physical NIC of a server is created. The method can also include intercepting another command to pair with a remote RDMA endpoint corresponding to the remote container. The intercepted another command contains data representing a routable network address of the remote RDMA endpoint in the RDMA computer network. Then, the RDMA endpoint created at the physical NIC of the server can be paired with the remote RDMA endpoint using the routable network address of the remote RDMA endpoint.

Abstract: A high dynamic range image sensors enabled by integrating broadband optical filters with individual sensor pixels of a pixel array. The broadband optical filters are formed of engineered micro or nanostructures that exhibit large differences in transmittance, e.g. up to 5 to 7 orders of magnitude. Such high transmittance difference can be achieved by using a single layer of individually designed filters, which show varied transmittance as a result of the distinct absorption of various material and structures. The high transmittance difference can also be achieved by controlling the polarization of light and using polarization-sensitive structures as filters. With the presence of properly designed integrated nanostructures, broadband transmission spectrum with transmittance spanning several orders of magnitude can be achieved. This enables design and manufacturing of image sensors with high dynamic range which is crucial for applications including autonomous driving and surveillance.

Abstract: A high dynamic range image sensors enabled by integrating broadband optical filters with individual sensor pixels of a pixel array. The broadband optical filters are formed of engineered micro or nanostructures that exhibit large differences in transmittance, e.g. up to 5 to 7 orders of magnitude. Such high transmittance difference can be achieved by using a single layer of individually designed filters, which show varied transmittance as a result of the distinct absorption of various material and structures. The high transmittance difference can also be achieved by controlling the polarization of light and using polarization-sensitive structures as filters. With the presence of properly designed integrated nanostructures, broadband transmission spectrum with transmittance spanning several orders of magnitude can be achieved. This enables design and manufacturing of image sensors with high dynamic range which is crucial for applications including autonomous driving and surveillance.

Abstract: A spectral sensing system includes an array of sampling optical elements (e.g. filters) and an array of optical sensors, and a deep learning model used to process output of the optical sensor array. The deep learning model is trained using a training dataset which includes, as training inputs, optical sensor output generated by shining each one of multiple different light waves with known spectra on the sampling optical element array for multiple different times, each time from a known angle of incidence, and as training labels, the known spectra of the multiple light waves. The trained deep learning model can be used to process output of the optical sensor array when light having an unknown spectrum is input on the sampling optical element array from unknown angles, to measure the spectrum of the light. The spectral sensing system can also be used to form a hyperspectral imaging system.

Abstract: Distributed storage systems, devices, and associated methods of data replication are disclosed herein. In one embodiment, a server in a distributed storage system is configured to write, with an RDMA enabled NIC, a block of data from a memory of the server to a memory at another server via an RDMA network. Upon completion of writing the block of data to the another server, the server can also send metadata representing a memory location and a data size of the written block of data in the memory of the another server via the RDMA network. The sent metadata is to be written into a memory location containing data representing a memory descriptor that is a part of a data structure representing a pre-posted work request configured to write a copy of the block of data from the another server to an additional server via the RDMA network.

Abstract: Distributed computing systems, devices, and associated methods of virtual RDMA switching are disclosed herein. In one embodiment, a method includes intercepting a command from an application in a container to establish an RDMA connection with a remote container on a virtual network. In response to the intercepted command, an RDMA endpoint at a physical NIC of a server is created. The method can also include intercepting another command to pair with a remote RDMA endpoint corresponding to the remote container. The intercepted another command contains data representing a routable network address of the remote RDMA endpoint in the RDMA computer network. Then, the RDMA endpoint created at the physical NIC of the server can be paired with the remote RDMA endpoint using the routable network address of the remote RDMA endpoint.

Abstract: A spectral sensing system includes an array of sampling optical elements (e.g. filters) and an array of optical sensors, and a deep learning model used to process output of the optical sensor array. The deep learning model is trained using a training dataset which includes, as training inputs, optical sensor output generated by shining each one of multiple different light waves with known spectra on the sampling optical element array for multiple different times, each time from a known angle of incidence, and as training labels, the known spectra of the multiple light waves. The trained deep learning model can be used to process output of the optical sensor array when light having an unknown spectrum is input on the sampling optical element array from unknown angles, to measure the spectrum of the light. The spectral sensing system can also be used to form a hyperspectral imaging system.

Abstract: Distributed computing systems, devices, and associated methods of virtual RDMA switching are disclosed herein. In one embodiment, a method includes intercepting a command from an application in a container to establish an RDMA connection with a remote container on a virtual network. In response to the intercepted command, an RDMA endpoint at a physical NIC of a server is created. The method can also include intercepting another command to pair with a remote RDMA endpoint corresponding to the remote container. The intercepted another command contains data representing a routable network address of the remote RDMA endpoint in the RDMA computer network. Then, the RDMA endpoint created at the physical NIC of the server can be paired with the remote RDMA endpoint using the routable network address of the remote RDMA endpoint.

Abstract: A high dynamic range image sensors enabled by integrating broadband optical filters with individual sensor pixels of a pixel array. The broadband optical filters are formed of engineered micro or nanostructures that exhibit large differences in transmittance, e.g. up to 5 to 7 orders of magnitude. Such high transmittance difference can be achieved by using a single layer of individually designed filters, which show varied transmittance as a result of the distinct absorption of various material and structures. The high transmittance difference can also be achieved by controlling the polarization of light and using polarization-sensitive structures as filters. With the presence of properly designed integrated nanostructures, broadband transmission spectrum with transmittance spanning several orders of magnitude can be achieved. This enables design and manufacturing of image sensors with high dynamic range which is crucial for applications including autonomous driving and surveillance.

Abstract: Optical sensing devices employing light intensity detectors integrated with nanostructures. In some embodiments, the nanostructures are 3D nanostructures having feature sizes in all three dimensions comparable to a wavelength range of the incident light, and are used for hyperspectral sensing. In some other embodiments, the nanostructures are simultaneous sensitive to both the spectrum and one or more of polarization, angle and phase information of the incident light field, to provide multi-modal optical sensing devices. In some other embodiments, each spatial pixel of an image sensor includes a group of sampling pixels configured for hyperspectral sensing and another group of sampling pixels configured for sensing polarization, angle or phase of the incident light.