Abstract: Embodiments of the present disclosure provide a method, a device, and a computer program product for edge resource aggregation. In one embodiment, a second edge device to which data associated with a first edge device will be transmitted is determined in an edge device cluster. Then, the data associated with the first edge device is transmitted to the second edge device. In this way, resources of multiple edge devices may be aggregated, thereby improving network efficiency and stability.

Abstract: Embodiments of the present disclosure relate to a method, a device, and a computer program product for data processing. The method comprises determining information associated with an attribute of a data packet from a source device; determining, based on the information, a target operation to be performed on the data packet; performing the target operation on the data packet; and causing the data packet on which the target operation has been performed to be sent to a target device of the data packet. In this manner, by using advantages of programmable switches, encapsulation and decapsulation operations are introduced for communications between edge Internet of Things network devices and data center applications, which can significantly increase the throughput.

Abstract: In illustrative embodiments, an information processing method, an electronic device, and a computer-readable storage medium are provided. The method includes: if it is determined that a predetermined indicator is detected in a header of a received packet, acquiring information for indicating a source port, a destination port, a forwarding indicator, a process identifier, and a maximum number of processes of the packet from the header; updating a port forwarding table based on the source port, the destination port, and the forwarding indicator; performing a user-defined action on the packet to obtain a result packet; and if it is determined that a quantity of process identifiers reaches the maximum number of processes, determining one or more forwarding ports for the result packet based on the maximum number of processes, the forwarding indicator, and the updated port forwarding table, to forward the result packet to the one or more forwarding ports.

Abstract: A method, an electronic device, and a computer-readable storage medium for information processing are provided according to example embodiments of the present disclosure. The method comprises receiving, at a storage device, a data block and fingerprint information correlated with the data block, the fingerprint information being configured to identify the data block; and determining a storage position of the received data block based on predetermined correlations between fingerprint information and storage positions and the received fingerprint information, the predetermined correlations comprising at least correlations between historical fingerprint information correlated with stored data blocks and historical storage positions. Thus, a storage position of a data block can be determined based on received fingerprint information and predetermined correlations between fingerprint information and storage positions, thereby improving efficiency of redundant data deletion.

Abstract: A method, an electronic device, and a computer-readable storage medium for information processing are provided according to example embodiments of the present disclosure. The method comprises receiving, at a storage device, a data block and fingerprint information correlated with the data block, the fingerprint information being configured to identify the data block; and determining a storage position of the received data block based on predetermined correlations between fingerprint information and storage positions and the received fingerprint information, the predetermined correlations comprising at least correlations between historical fingerprint information correlated with stored data blocks and historical storage positions. Thus, a storage position of a data block can be determined based on received fingerprint information and predetermined correlations between fingerprint information and storage positions, thereby improving efficiency of redundant data deletion.

Abstract: Embodiments of the present disclosure relate to a method, a device, and a computer program product for processing data. The method includes: loading, at a switch and in response to receipt of a model loading request from a terminal device, a data processing model specified in the model loading request. The method further includes: acquiring model parameters of the data processing model from the terminal device. The method further includes: processing, in response to receipt of to-be-processed data from the terminal device, the data using the data processing model based on the model parameters. Through the method, data may be processed at a switch, which improves the efficiency of data processing and the utilization rate of computing resources, and reduces the delay of data processing.

Abstract: Embodiments of the present disclosure relate to a method, a device, and a computer program product for data processing. The method comprises determining information associated with an attribute of a data packet from a source device; determining, based on the information, a target operation to be performed on the data packet; performing the target operation on the data packet; and causing the data packet on which the target operation has been performed to be sent to a target device of the data packet. In this manner, by using advantages of programmable switches, encapsulation and decapsulation operations are introduced for communications between edge Internet of Things network devices and data center applications, which can significantly increase the throughput.

Abstract: In illustrative embodiments, an information processing method, an electronic device, and a computer-readable storage medium are provided. The method includes: if it is determined that a predetermined indicator is detected in a header of a received packet, acquiring information for indicating a source port, a destination port, a forwarding indicator, a process identifier, and a maximum number of processes of the packet from the header; updating a port forwarding table based on the source port, the destination port, and the forwarding indicator; performing a user-defined action on the packet to obtain a result packet; and if it is determined that a quantity of process identifiers reaches the maximum number of processes, determining one or more forwarding ports for the result packet based on the maximum number of processes, the forwarding indicator, and the updated port forwarding table, to forward the result packet to the one or more forwarding ports.

Abstract: Embodiments of the present disclosure relate to a method for encryption and decryption, a programmable switch, and a computer program product. The method comprises receiving, at a programmable switch, encrypted data to be sent to a certain Internet of Things (IoT) device, wherein the encrypted data is encrypted using a public key of the programmable switch. The method further comprises decrypting the encrypted data using a private key of the programmable switch to obtain decrypted data, and then sending the decrypted data from the programmable switch to the IoT device. According to the embodiments of the present disclosure, the encryption and decryption operations of the IoT device can be transferred to the programmable switch for processing, and the programmable switch is used to help the IoT device perform encryption and decryption.

Abstract: Techniques are disclosed for expanding a memory of a graphics processing unit. For example, a method comprises: receiving, from a graphics processing unit coupled to a host, an indication that data to be accessed by the graphics processing unit is absent in the memory of the graphics processing unit, where the data is stored in a storage device coupled to the host; in response to receiving the indication, determining, from a plurality of candidate paths, a path for migrating the data from the storage device to the memory of the graphics processing unit; causing the data to be migrated from the storage device to the memory of the graphics processing unit via the determined path; and instructing the graphics processing unit to access the data from the memory of the graphics processing unit. As such, a memory of a GPU can be expanded with an external storage device.

Abstract: Techniques are disclosed for expanding a memory of a graphics processing unit. For example, a method comprises: receiving, from a graphics processing unit coupled to a host, an indication that data to be accessed by the graphics processing unit is absent in the memory of the graphics processing unit, where the data is stored in a storage device coupled to the host; in response to receiving the indication, determining, from a plurality of candidate paths, a path for migrating the data from the storage device to the memory of the graphics processing unit; causing the data to be migrated from the storage device to the memory of the graphics processing unit via the determined path; and instructing the graphics processing unit to access the data from the memory of the graphics processing unit. As such, a memory of a GPU can be expanded with an external storage device.

Abstract: Embodiments of the present invention disclose a virtual network device. The virtual network device is configured to implement a broadband network gateway BNG function, and the virtual network device includes a load balance unit and a first forwarding unit; the load balance unit is configured to: receive a first data packet from a first user terminal, determine the first forwarding unit according to the first data packet and a first correspondence, and send the first data packet to the first forwarding unit, where the first correspondence indicates mapping from the first user terminal to the first forwarding unit; and the first forwarding unit is configured to receive and forward the first data packet. In addition, the embodiments of the present invention further disclose a method for implementing load sharing in a virtual network device.

Abstract: Embodiments of the present invention disclose a virtual network device. The virtual network device is configured to implement a broadband network gateway BNG function, and the virtual network device includes a load balance unit and a first forwarding unit; the load balance unit is configured to: receive a first data packet from a first user terminal, determine the first forwarding unit according to the first data packet and a first correspondence, and send the first data packet to the first forwarding unit, where the first correspondence indicates mapping from the first user terminal to the first forwarding unit; and the first forwarding unit is configured to receive and forward the first data packet. In addition, the embodiments of the present invention further disclose a method for implementing load sharing in a virtual network device.

Abstract: The present invention provides novel methods of accurately and efficiently reconstructing parameter maps in MRI data. In certain embodiments, the methods have reduced data acquisition time or improved spatial resolution when compared to methods standard in the art. In other embodiments, the methods have reduced acquisition and reconstruction time when compared to other acceleration methods in the art in parametric mapping. In other embodiments, the methods are less susceptible to influence of eddy currents when compared to other acceleration methods in the art in parametric mapping.

Abstract: Embodiments of the present invention disclose a virtual network device. The virtual network device is configured to implement a broadband network gateway BNG function, and the virtual network device includes a load balance unit and a first forwarding unit; the load balance unit is configured to: receive a first data packet from a first user terminal, determine the first forwarding unit according to the first data packet and a first correspondence, and send the first data packet to the first forwarding unit, where the first correspondence indicates mapping from the first user terminal to the first forwarding unit; and the first forwarding unit is configured to receive and forward the first data packet. In addition, the embodiments of the present invention further disclose a method for implementing load sharing in a virtual network device.

Abstract: The present invention provides novel methods of accurately and efficiently reconstructing parameter maps in MRI data. In certain embodiments, the methods have reduced data acquisition time or improved spatial resolution when compared to methods standard in the art. In other embodiments, the methods have reduced acquisition and reconstruction time when compared to other acceleration methods in the art in parametric mapping. In other embodiments, the methods are less susceptible to influence of eddy currents when compared to other acceleration methods in the art in parametric mapping.