203 results about "Networking hardware" patented technology

A fiber-optic network that uses as a hybrid telecommunication and sensing system when telecommunication signals and probe signals are transmitted across a shared fiber strand in either a co-propagated or counter-propagated direction is disclosed. Probe signals generated by a sensing termination system and / or by one or more end devices are used to analyze conditions affecting network hardware and / or events occurring within the fiber distribution area.

The invention discloses a neural network extension execution computer system based on RISC-V. The neural network extension execution computer system comprises an RISC-V processor based on an out-of-order execution system structure, a neural network hardware accelerator, an instruction memory, a data memory and a dynamic scheduler. The dynamic scheduler is used for switching different working modes for the computer system according to the current calculation load, and distributing a calculation task to be executed to the RISC-V processor or the neural network hardware accelerator for execution; the working modes comprise a standard calculation mode, a neural network acceleration mode and a low power consumption mode; the computing tasks comprise a general computing task and a neural network computing task. According to the technical scheme, the intelligence of the computer system is improved, and the calculation efficiency, the resource consumption and the power consumption are well balanced.

The invention provides a system and a method for generating CAN (Controller Area Network) mailbox configuration and data routing configuration based on a communication matrix table. The method comprises the following steps of: reading a predefined finished automobile CAN communication network signal routing information table; analyzing the whole vehicle CAN communication network signal routing information table to obtain CAN communication configuration parameters of each CAN channel; the number of receiving mailboxes and the number of sending mailboxes of each CAN channel are distributed according to the maximum number of receiving mailboxes and the maximum number of sending mailboxes supported by the whole vehicle CAN communication network hardware for each CAN channel and CAN communication configuration parameters of each CAN channel; and when the number of the messages to be processed of a certain CAN channel exceeds the allocated mailbox number, classifying the messages, and multiplexing the same mailbox by the messages belonging to the same class. According to the method, CAN mailbox resources are used to the greatest extent, and the reliability of data routing is ensured; when the whole vehicle communication network signal routing table is frequently changed, the reliability of the routing function is ensured, and meanwhile, the development efficiency is improved.

The invention belongs to the technical field of embedded hardware acceleration and computer vision, and relates to a quantitative lightweight neural network hardware accelerator and a design method thereof. The hardware accelerator comprises a top-layer control module, a memory control module, a reconfigurable memory, an input data preprocessing module, a core computing unit and a post-processing module; the memory control module controls the reconfigurable memory to send data information to the top-layer control module, the input data preprocessing module and the post-processing module; the input data preprocessing module is used for data and weight preparation; and the core calculation unit is used for completing data preparation, weight preparation and post-processing through a multiplier, an adder tree and an accumulator array. The hardware accelerator has higher computing performance, hardware resource utilization rate and data reuse rate; according to the method, the problem of low storage data utilization rate of the input and output cache is solved; hardware operator fusion of a convolution operator and a pooling operator is realized, and storage delay and access power consumption of data reading and writing are reduced.

The invention provides a large-scale constellation network semi-physical simulation test system and method, the system comprises a constellation simulator, a satellite-borne router, ground detection equipment and a protocol conversion network bridge, a Mininet simulation platform is deployed, and dynamic topology adjustment, an OSPF dynamic routing protocol and a multi-scene test function are supported; a user can observe network topology change, route updating time and key performance indexes in real time. A semi-physical simulation architecture is realized through unified IP subnet interconnection and the same routing protocol, a space-time synchronization engine ensures that virtual and real node routing tables take effect at an appointed moment, and in combination with a protocol consistency sandbox, hidden faults caused by inconsistent behaviors of hardware and software protocol stacks are solved, and the service life of the virtual and real node routing tables is prolonged. A plurality of real nodes are accessed in a large-scale network environment, and end-to-end data transmission test, security policy verification and protocol compatibility analysis are carried out; the invention provides an efficient, flexible and reliable solution for the design, verification and optimization of the satellite network, and has a wide application prospect.

The invention relates to the field of computer hardware virtualization, and discloses a CAN network virtualization method, system, device and medium, a virtual CAN network is established through QEMU virtual machines, each QEMU virtual machine can be used as a node, all the nodes are connected through a Kvaser Virtual CAN Driver driver to form a virtual CAN network, the virtual CAN network can be used for simulation and testing, and the virtual CAN network can be used for simulating and testing. Therefore, logic analysis of the virtual device is separated from actual operation of the device, the front end is mainly used for reading, writing and the like of a register, and when hardware operation is involved, the operation is transmitted to the rear end, so that coupling is reduced, the flexibility of a model is improved, and the reliability of the model is improved. The problem that drivers and application software cannot be developed and verified in advance due to high CAN network hardware cost and long development period is solved.

The invention discloses a digital substation electric power industrial control system network hardware asset identification method and device, relates to the technical field of electric power industrial control system asset identification, and aims to identify relevance between to-be-accessed target hardware assets and existing assets by calculating similarity and update a hardware asset library in time. False report and confusion caused by network change are reduced, and efficiency and accuracy of network management are improved. The method comprises the following steps: constructing a hardware asset attribute index; preprocessing the hardware asset data of each hardware asset according to the hardware asset attribute indexes, and constructing a hardware asset library; in response to the hardware asset access request, obtaining a to-be-accessed target hardware asset, preprocessing target hardware asset data of the target hardware asset to obtain a target attribute sequence, and calculating the similarity between the target hardware asset and each hardware asset by using the target attribute sequence; and updating the hardware asset library by referring to the similarity between the target hardware asset and each hardware asset.

The application discloses a medical image transmission method based on a 5G network, and specifically comprises the following steps: step one, hardware architecture; step two, image acquisition and encapsulation; step three, decapsulation and restoration; and step four, fault-tolerant processing, and relates to the technical field of medical communication. The medical image transmission method based on the 5G network constructs a 5G network hardware framework system, utilizes the low-latency and high-speed characteristics of the 5G network for information transmission, and cooperates with a client, a slice information unit, a state information monitoring unit, a fault-tolerant information unit and an identification unit to form a client message system, utilizes a server, a transmission data record module, a slice image information module, a data information module and a data integration module to form a server end heat preservation system, and after encapsulation and decapsulation, the medical image with a large amount of data can be completely and quickly transmitted and saved to a remote secure server while ensuring the high-speed stability of the transmission environment.

One or more computing devices, systems, and / or methods for visual troubleshooting a network device setup. Images of the network device setup are provided to the system. A GenAI component processes the images to generate one or more device identifying features. The features are further processed to identify the device. The system utilizes hardware-specific information to prompt the GenAI component to answer troubleshooting-related questions concerning the device setup. The images may be pre-processed to include one or more visual guides to assist the GenAI component.

The present invention proposes a method for realizing in-memory computing based on a small-signal ferroelectric capacitor with non-destructive readout, belonging to the field of novel computing architectures. The present invention realizes the linearly inseparable comparison operation of a content-addressable memory (CAM) unit and the local multiplication operation of a signed synaptic unit on a ferroelectric capacitor, reducing the hardware cost. Compared with the method of distance measurement based on the traditional CAM architecture, it has higher computational linearity and detection range, as well as lower search power consumption; at the same time, compared with the resistive synaptic array, it has no DC power consumption and no potential DC conduction path, avoiding the "IR drop" problem of large-scale arrays, and has lower computational power consumption, showing significant advantages in edge machine learning tasks based on feature retrieval and neural network hardware acceleration.

The invention relates to a credential cross-platform unidirectional optical isolation data transmission and release method and system, and belongs to the field of data processing. The method comprises the following steps: triggering a signal-creative cross-platform one-way data acquisition isolation optical shutter through an initialization protocol, the one-way data acquisition isolation optical shutter covering an intranet processing layer, an extranet processing layer and a single link physical optical fiber layer, and executing the initialization protocol to perform dual-channel independent initialization on network hardware interfaces of the intranet processing layer and the extranet processing layer; the user configuration parameters are analyzed, and a configuration strategy is generated based on the user configuration parameters and used for configuring a one-way data acquisition isolation optical shutter; generating and executing an interface activation protocol of the one-way data acquisition isolation optical shutter based on the user configuration parameters; the intranet processing layer of the unidirectional data acquisition isolation optical shutter receives field network data and unidirectionally transmits the field network data to the extranet processing layer, and the extranet processing layer delivers the field network data to a specified server. According to the invention, one-way secure transmission of data between the internal network and the external network is realized, and no information flowing to the internal network from the external network is ensured.

A fiber-optic network that uses as a hybrid telecommunication and sensing system when telecommunication signals and probe signals are transmitted across a shared fiber strand in either a co-propagated or counter-propagated direction is disclosed. Probe signals generated by a sensing termination system and / or by one or more end devices are used to analyze conditions affecting network hardware and / or events occurring within the fiber distribution area.

The methods of communicating audio data disclosed herein decouple the generating audio data by an audio source from sending the audio data over a network cloud by a sending side network hardware abstraction layer by using a thread pool. The thread pool decouples the generation of the audio data from the sending of the audio data thereby preventing network delays from impacting audio data generation. The methods decouple the receiving of the audio data from the network cloud by pulling the audio data through at least portions of the receiving side thereby delivering precise quantities of audio data at precise times to an audio device. The data requirements of the audio device control the flow of audio data through the receiving side, not the receiving of audio data from the network cloud.

The invention discloses an Ethernet network hardware accelerator control method and device, equipment and a medium, and the method comprises the steps: responding to an operation system dormancy or wake-up operation, writing mark information into a file node corresponding to an Ethernet network hardware accelerator through a preset process in a user mode, the mark information comprising a first preset value and a second preset value, the first preset value represents dormancy of the operating system, and the second preset value represents awakening of the operating system; and opening or closing the Ethernet network hardware accelerator based on the mark information through a network hardware accelerator drive in a kernel mode. Therefore, when the operating system is dormant or awakened, the preset process in the user mode and the network hardware accelerator in the kernel mode drive the file node based on the Ethernet network hardware accelerator to interact, so that the Ethernet network hardware accelerator is opened or closed, the reliability of the Ethernet network hardware accelerator can be guaranteed, and the reliability of the Ethernet network hardware accelerator is improved. Therefore, the communication rate of the Ethernet is improved, the overhead of a CPU (Central Processing Unit) is reduced, and the user experience is improved.

The invention discloses a target detection-oriented lightweight network hardware acceleration method and a target detection method, and the method comprises the steps: employing a lightweight module to improve a YOLO network, and obtaining a target detection network; sparse processing is carried out on the scaling coefficient of the BN layer in the target detection network to obtain a BN layer sparse factor; determining a module of which the parameter quantity exceeds a threshold value in the target detection network according to the BN layer sparse factor, and performing pruning operation to obtain a pruned target detection network; dividing the pruned target detection network into a convolution accelerator module, a depth separable convolution accelerator module, a pooling accelerator module and an up-sampling accelerator module; and selecting an acceleration algorithm according to a Roofline model, and optimizing the convolution accelerator module, the depth separable convolution accelerator module, the pooling accelerator module and the up-sampling accelerator module to obtain an optimized target detection network. According to the invention, the problems of poor real-time performance and overhigh power consumption in target detection are solved.

A system and method is provided for designing and optimizing hardware accelerators for neural networks. During a pre-design phase, rules are extracted from compilation patterns that describe conversion between neural network operators, coarse-grained operators, and fine-grained dataflow. A fast mapper for converting neural network models to coarse-grained operator descriptions and a dataflow mapper are generated. A coarse-grained design phase employs an architecture optimizer to generate plural provisional hardware accelerator designs The coarse-grained operator descriptions are simulated using a coarse-grained simulator to obtain performance metrics of each provisional accelerator design. A fine-grained design phase employs a dataflow mapper and fine-grained simulator to finalize provisional hardware accelerator designs. A hardware accelerator is generated from a finalized hardware accelerator design and a corresponding software toolchain is created including a compiler and software development kit (SDK) for programming, debugging, and deploying the hardware accelerator design.

The invention discloses a network configuration method and device, electronic equipment and a storage medium, relates to the technical field of container cluster management, realizes network multiple models of a distributed cluster by constructing a multi-network model plug-in integrated with multiple functions, and solves the technical problem that the cluster needs multiple network plug-ins, so that the multi-network plug-ins cannot be used in the distributed cluster. The complexity of cluster deployment and maintenance is high; multiple network plug-ins work cooperatively, the network problem boundary is not clear, and the problem checking complexity is high; the multi-network plug-in lacks a unified IPAM management mechanism; only the setting of a workload granularity network model is supported, the implementation scheme of Namespace granularity is lacked, and the usability is poor; due to different network hardware resources required by service applications, a user needs to customize scheduling nodes and the like of a workload; the application performance of the distributed cluster is improved, through configuration of the network resources, the network configuration process is simplified, the network configuration efficiency is improved, and the application performance of the distributed cluster is further improved.

Systems and methods are provided for generating libraries for hardware realization of neural networks. The method includes obtaining a plurality of neural network topologies. Each neural network topology corresponds to a respective neural network. The method also includes transforming each neural network topology to a respective equivalent analog network of analog components. The method also includes generating a plurality of lithographic masks for fabricating a plurality of circuits. Each circuit implements a respective equivalent analog network of analog components.

A machine learning-based proxy model may generate measurements of quality of user experience in a vehicle-based communication network in the absence of direct feedback from the users regarding the user experience. Once generated, the proxy model may be applied to observed operational parameters of the on-board network to quantify the user experience for any user in any given instance. User experience measurements (e.g., trends identified therein) may be utilized to identify and implement adjustments to hardware, firmware, software, and / or service procedures associated with implementation of the on-board network. These adjustments may be implemented between transits of the vehicle, or in some cases, during transit of the vehicle to improve the user experience over the duration of use of the vehicle-based communication network.

Systems and methods herein are for at least one circuit that can determine that a network reference associated with a first network hardware is subject to a network performance degradation in a network, can cause suspension of traffic flow associated with the network reference, can save configuration for at least the network reference at a node associated with the first network hardware, and can cause the configuration to be deployed in a second network hardware so that the network reference that was previously in the first network hardware is provided from the second network hardware to resume the traffic flow.

A cluster configuration request to form a hyperconverged computing infrastructure (HCI) cluster in a cloud computing environment is processed. Based on the cluster configuration request and any other cluster specifications, a plurality of bare metal computing nodes of the cloud computing environment are configured to operate as an HCI cluster. First, a tenant-specific secure network overlay is formed on a first set of tenant-specific networking hardware resources. Then, the tenant-specific secure network overlay is used by an orchestrator to provision a second set of tenant-specific networking hardware resources. The second set of tenant-specific networking hardware resources are configured to interconnect node-local storage devices into a shared storage pool having a contiguous address space. Top-of-rack switches are configured to form a network overlay on the first set of tenant-specific networking hardware resources. Then, top-of-rack switches are configured to form a layer-2 subnet on the second set of tenant-specific networking hardware resources.

System and method that enable a userspace compartmented IP Stack and network interfaces between a user device and communications devices associated with the user device, including mobile devices. The system and method include userspace bridging of data to / from external communications devices and applications onboard the user device. The system and method include userspace control of the network interfaces, routing between disparate IP networks, and routing of external IP network interfaces into user device interfaces. The system and method includes incorporating userspace device drivers required for external network hardware interface.

Disclosed is a system for testing batteries, comprising: a battery tester comprising: a tester scanner or camera for capturing an obtained battery identifier; a tester network hardware for transmitting the obtained battery identifier; and a server comprising: a database having data, the data comprising at least one historic battery identifier and associated historic battery characteristic and configured to compare the data with the obtained battery identifier; wherein the battery tester is configured to capture a battery identifier from a battery and transmit the battery identifier to the server.

The application discloses a transconductance variable field effect transistor array suitable for a dendritic network hardware and application, and belongs to the technical field of semiconductor integrated circuits.The application realizes three-element multiplication of a storage variable and two input variables based on a single transconductance variable field effect transistor, and realizes mapping of a dendritic network core algorithm based on a complementary device array.Compared with a traditional neural network hardware which realizes nonlinear transformation by using a neuron activation circuit, the application realizes nonlinear transformation by using intrinsic nonlinearity of a device, effectively reduces design complexity, optimizes area and power consumption of a system peripheral circuit, and has important significance for design of a high-performance artificial intelligence computing system.

The utility model discloses a computer network hardware equipment installation device which comprises a case, case seats are installed on the upper side and the lower side of the case, a cushion seat is installed below the case seats, a lower seat is installed below the cushion seat, a heat removal groove is formed in the surface of the lower seat, special-shaped grooves are formed in the two sides of the heat removal groove, and the heat removal groove and the special-shaped grooves are communicated with each other. A base plate is embedded in the special-shaped groove, and a supporting plate is arranged on one side of the base plate. The pedestal is installed at the bottom of the case and divided into an upper part and a lower part, the lower part is a lower seat, and the lower seat and the pedestal are movably connected through sleeving connection between the hollow sleeve and the sleeve rod, so that the pedestal and the lower seat are conveniently separated, the space of the bottom of the pedestal is enlarged, the heat dissipation space is increased, and better heat dissipation is facilitated; a special-shaped groove of a T-shaped structure is formed in the surface of the cushion seat, the cushion plate and the supporting plate are moved, and the cushion plate is used for increasing the height of the cushion seat, so that the space between the cushion seat and the lower seat is increased.

A black screen control system for nitric acid production, comprising network hardware nodes deployed in a local area network and control units running on the network hardware nodes, the control units are independently deployed at each operation station and adopt a multi-thread isolation architecture. The invention allocates time-consuming bottom layer data collection, processing and front-end UI interaction to independent threads for asynchronous execution, ensuring the ultimate smoothness and stability of the system under high concurrency; the state control and abnormal response are separated, the system enters the energy-saving black screen state under normal conditions to reduce power consumption, and when an abnormal signal occurs, the system is instantly awakened across threads and a safety alarm is triggered, realizing true safe unattended operation; the control units are independently deployed and run at each operation station, and are uniformly scheduled by a master control module, avoiding the whole line paralysis caused by single point failure or network fluctuation, and ensuring the absolute independence and reliability of a single device.

Respective network addresses are assigned to a networking hardware device and a network function accelerator incorporated within an offloading card of a virtualization server by a control plane server of a virtualized computing service. A compute instance is launched at the virtualization server using a virtualization controller of the offloading card. The compute instance executes a network function of a radio-based application in response to a request received using the network address assigned to the hardware device, and requests execution of a second network function at the accelerator. The result of the second network function is sent to a radio unit of the application using the address assigned to the accelerator.

The invention provides an optimal phase shift parameter design method based on a switch multiplexing type multi-resonant switched capacitor, and the method comprises the steps: determining a control parameter combination according to the value interval of a preset phase shift ratio and a preset frequency ratio; determining resonance network hardware parameters according to the preset target equalization power and the preset resonance frequency; according to a preset soft switching constraint condition, verifying a parameter set composed of all control parameter combinations and resonance network hardware parameters; if the parameter set meets the preset soft switching constraint condition, taking the parameter set as an element of a feasible solution set; screening a parameter set with the minimum root mean square value of the resonance current according to the feasible solution set; and according to the parameter set with the minimum root-mean-square value of the resonance current, screening the parameter set with the minimum peak value of the resonance current as an optimal working point. According to the invention, through collaborative optimization of multiple performance indexes, on the premise of following the soft switching constraint, the switching loss and the operation loss of the switching tube are reduced, and the overall efficiency and the reliability of the equalizer are improved.

Respective network addresses are assigned to a networking hardware device and a network function accelerator incorporated within an offloading card of a virtualization server by a control plane server of a virtualized computing service. A compute instance is launched at the virtualization server using a virtualization controller of the offloading card. The compute instance executes a network function of a radio-based application in response to a request received using the network address assigned to the hardware device, and requests execution of a second network function at the accelerator. The result of the second network function is sent to a radio unit of the application using the address assigned to the accelerator.