54 results about "Wire speed" patented technology

The invention relates to a multi-tenant unified storage routing table item compression method and system based on identification mapping. The method comprises the following steps: in a routing table entry insertion stage, dynamically distributing a minimum available reduce ID for each new tenant, and ensuring that an IP prefix space of the new tenant is not overlapped with a tenant using the same reduce ID; in the lookup stage, a message VNI is firstly converted into a reduce ID through a mapping table realized based on an RAM, then the reduce ID and a destination IP are used as a joint key to look up the TCAM, and by constructing a mapping relation from the VNI to the reduce ID, the table item bit width of the TCAM is remarkably reduced on the premise of ensuring the routing correctness. The method is suitable for an IPv4 / IPv6 multi-tenant environment, the hardware area and power consumption can be effectively reduced especially in an FPGA or ASIC high-performance data plane, and meanwhile the line speed forwarding capacity is kept.

The utility model discloses a high -speed spark -discharge wire cutting device with wire storage mechanism is provided in the electric spark wire cutting technical field, including the first wire storage cylinder and second wire storage cylinder of vertical setting in the wire storage box. The high -speed spark -discharge wire cutting device with wire storage mechanism, through the first wire storage cylinder and second wire storage cylinder of vertical arrangement, cooperate the sliding block of sliding setting, realize the conductive wire on the first wire storage cylinder after cutting section and set up to second wire storage cylinder, thereby realize single -layer wire storage, and the sum of the conductive wire length between first fixed pulley and first movable pulley and the conductive wire length between second fixed pulley and second movable pulley remain unchanged in the moving process of moving block, thereby realized the conductive wire moving stability, compared with existing device, avoided the problem that the wire speed changed when the electrode wire was arranged in the laminated arrangement on the wire cylinder, improved wire cutting stability.

The invention relates to a dynamic purification method for 5G user plane traffic. The method comprises the following steps: receiving mixed user plane traffic in a network, preprocessing the received traffic, analyzing the preprocessed traffic according to a message header feature and an extension structure of a user plane bearer protocol, and dynamically identifying and separating out 5G user plane data stream; performing application layer depth detection on the identified 5G user plane data stream, identifying a second type of service data stream established based on the session control signaling, and analyzing a dynamic transmission channel parameter of the second type of service data stream; generating a flow filtering strategy according to the analyzed dynamic transmission channel parameters; and executing a dynamic strategy to carry out linear speed filtering on the 5G user plane data stream. The 5G user plane data flow is identified by analyzing the protocol head feature and the extension structure in the user plane flow in real time, and deep detection is performed on the application layer, so that the 5G flow can be quickly screened in a high-speed network environment.

This application provides a flow control method, an artificial intelligence distributed training method, an electronic device, a storage medium, and a program product. The method includes: acquiring traffic characteristic information of a data stream to be transmitted sent by an application layer program; generating corresponding flow table rules based on the traffic characteristic information; pre-configuring the flow table rules to a network device and controlling the flow table rules to be in an inactive dormant state; upon receiving a trigger signal sent by the application layer program, sending a control command to the network device to switch the flow table rules to an active state, so that the data stream to be transmitted is transmitted via a target path based on the active flow table rules; the trigger signal is used to instruct the data stream to be transmitted to start transmission. This avoids the problem of repeated PFC oscillations caused by switch buffer congestion due to sudden large traffic flows in traditional RoCE networks, enabling large data streams to be transmitted at a stable rate close to line speed.

The invention relates to a winding type precise coating processing device for a multi-channel flexible implanted wire. The device comprises a multi-coil pay-off mechanism, a wrapping head, a curing unit, a desolventizing unit, a drying unit, an online metering unit and a controller. A wrapping head is provided with an independent main guide belt and a soluble phase guide wire channel, and the main guide belt and the soluble phase guide wire channel are wound on the periphery of the core material in the same direction and at the same pitch when each layer of wrapping is formed; the curing unit adopts a one-time curing and final curing segmentation mode, the screw pitch and the geometric phase are locked firstly, and then the final morphology is frozen; the desolventizing unit is used for dissolving or extracting the soluble guide wire after final curing, so that residues are avoided; the on-line metering unit collects a screw pitch signal in real time, the controller carries out power spectrum analysis, and a beat frequency spectrum peak amplitude is used as a feedback closed loop to adjust the line speed and the wrapping head angular speed. According to the scheme, the beat frequency effect caused by the multi-layer pitch differential is effectively inhibited, the geometric stability of the wire is ensured, the flexibility and the insulating property are improved, no residue is ensured, and the consistency and the long-term reliability of the flexible implanted wire are remarkably improved.

A method to create a serial wire speed firewall that can monitor and enforce security policy on a CAN buy network and prevent cyber-attacks.

This application relates to a silicon wafer cutting method and apparatus. The silicon wafer cutting method includes the following steps: setting a brand-new cutting wire on a main roller mechanism, driving the cutting wire in the cutting wire mesh to run cyclically through the main roller mechanism, cutting the silicon rod entering the cutting wire mesh into silicon wafers. The cyclical running includes an initial forward run, a reverse run, and a second forward run. The silicon wafer cutting method of this application, by making the cutting wire run cyclically during the cutting process, utilizes a cutting wire three times in a reasonable manner. In the three utilization processes, the cutting depth is maximized during the initial forward run with the highest wire speed, and the cutting depth is minimized during the second forward run with the lowest wire speed. This effectively controls the cutting wire speed and cutting depth according to the wear condition of the cutting wire, so as to make full use of the cutting wire while ensuring the normal operation of the wire bow, and thus ensuring the cutting quality and the quality of the silicon wafer.

Techniques for utilizing a portion of a communication session identifier (e.g., a Session-ID, an SPI, a CID, a DCID, and / or the like) to indicate a target routing device (e.g., a VPN and / or ZTNA termination device) for establishing control plane session(s) and / or data plane session(s) at wire-speed in a networked computing environment. The routing device(s) of a networked computing environment may generate a communication session identifier and send the communication session identifier to the client device, such that subsequent packets send from the client device may be forwarded to the proper routing device indicated by the communication session identifier for establishment of one or more data plane sessions. Additionally, data plane sessions may be established using a Resumed Handshake rather than a full handshake that is typically required, as Session Resumption utilizes the assigned communication session identifier for mapping.

The invention discloses a production method of a flat cable, which comprises the following steps of: S1, conductor preparation: performing back-twist twisting on a conductor wire to prepare a guide wire bundle, the back-twist rate being 100%, and the pitch being (11-13) * OD + / -1mm; s2, insulating extrusion: starting an extruder to heat and pull a guide wire bundle, and enabling the preheated guide wire bundle to pass through the extruder so as to extrude and coat an insulating layer on the surface of the guide wire bundle to form a cable; when the tow is preheated, the preheating temperature is set to be T1 based on the moving linear speed of the tow; and S3, post-processing: performing water-cooling shaping on the cable to keep the shape, wherein the water-cooling temperature is 12-18 DEG C. By heating the conductor and optimizing the synergistic effect of the insulating rubber material formula, the cable wire can simultaneously meet the requirements of zero hoisting displacement and no copper wire breakage during wire stripping; under the condition of high hoisting weight, enough adhesive force exists between the conductor and the insulating layer; and during the wire stripping operation, the adhesive force between the conductor and the insulating layer is ensured to be small, and the processing adaptability and the use reliability of the cable are improved.

The application discloses an integrated tinning machine intelligent control method and system combined with data fusion, and belongs to the field of automatic control. The method comprises the following steps: a multi-dimensional data fusion sensing step, in which the copper-clad steel wire speed, acceleration, width, yaw displacement and rectifier feedback current are acquired; a dynamic current density model construction step, in which a theoretical current reference value is output based on the Faraday electrolysis law and real-time bath parameters; a transient feedforward compensation control step, in which a current compensation amount is calculated according to the acceleration and phase delay time constant, and a predictive current control instruction is generated; and a side effect cooperative optimization step, in which the anode plate position and shielding state are dynamically adjusted according to the width and yaw displacement. The application is suitable for complex production conditions and improves the automatic operation level of the tinning machine set.

This invention discloses a programmable data plane high-intensity traffic response method and system based on feature distribution, belonging to the field of network traffic classification technology. The programmable data plane high-intensity traffic response method based on feature distribution includes: extracting features from each flow in the data plane and determining whether the features are matched by the current rule; if matched, processing according to the corresponding rule; if not matched, processing as an outlier and recording it; periodically sampling processed flows, triggering rule updates when outliers reach a preset threshold; the control plane uses a density-based clustering algorithm to cluster sampled points to obtain the current traffic feature distribution shape, extracting the boundaries of each cluster to form a high-dimensional rectangle as a new rule, and formulating corresponding processing measures based on spatial features and meta-features, and issuing them to the data plane. This invention achieves line-rate processing and dynamic adaptation on resource-constrained programmable hardware, effectively addressing feature drift under high-intensity traffic while ensuring interpretability.

The invention relates to the technical field of network communication, in particular to a programmable data packet scheduling method and system for a network switching chip, and the method comprises the steps: constructing a hierarchical scheduling tree; the hierarchical scheduling tree comprises a leaf node level and a root node level, and each node is provided with a bitmap mapping bucket queue, a dynamic priority window mapping module and a reconfigurable priority calculator. According to the method, the technical problem that an existing scheduler is insufficient in flexibility, performance and expansibility can be solved, and then multi-algorithm dynamic deployment, priority range expansion and hierarchical scheduling optimization are achieved while the linear speed processing performance is kept.

The application relates to a multi-tenant uniform storage routing table item compression method and system based on identification mapping. The method comprises the following steps: in a routing table item insertion stage, a minimum available reduceID is dynamically allocated for each new tenant, and it is ensured that the IP prefix space of the tenant is not overlapped with a tenant using the same reduceID; in a lookup stage, a message VNI is converted into a reduceID through a mapping table based on RAM, and then reduceID+destination IP is used as a joint key to query a TCAM, and the mapping relationship between the VNI and the reduceID is constructed, so that the bit width of the TCAM table item is significantly reduced under the premise of ensuring routing correctness. The application is suitable for an IPv4 / IPv6 multi-tenant environment, and can effectively reduce the hardware area and power consumption in a FPGA or ASIC high-performance data plane, while maintaining the line-speed forwarding capability.

A cloud server obtains a defined signature from a first communication device, where the defined signature is associated with a digital asset at the first communication device. The cloud server validates the defined signature based on a search of the defined signature in a main action database in the cloud server. The cloud server determines that a copyright check is required for the digital asset associated with the defined signature. The cloud server determines an occurrence of copyright violation for digital asset based on resemblance search of the digital asset in copyright database. The cloud server interrupts, based on the validation of the defined signature and the occurrence of the copyright violation, consumption of one or more data packets of the digital asset from being consumed at a datalink layer and a wire speed, prior to exposure of the one or more data packets to an application layer of the first communication device.

The application relates to the technical field of network traffic processing, and discloses a cross-message HTTP domain name matching method and system on a network distribution device. The method comprises the following steps: receiving and analyzing a TCP segment message, obtaining payload data and quintuple information, and then performing first domain name retrieval through preset head and tail anchor points. If the head and tail anchor points are simultaneously identified in a single message, the domain name information is directly extracted to complete matching. If only one of the anchor points is identified, the message is marked as a to-be-recombined message and is sent to a TCP segment recombination step. During recombination, the recombination table item of the FPGA integrated HBM is accessed based on the quintuple hash value, the payload data is written into the corresponding position, the recombined payload data is output when a preset output condition is met, and finally, second domain name retrieval is performed on the recombined data. After the head and tail anchor points are identified, the domain name is extracted to complete matching. The method realizes accurate matching of cross-message domain names under a 400G large bandwidth, takes into account single-message line-speed processing performance, solves the problem of cross-message domain name missing matching, and is suitable for network environments in which multiple websites share an IP.

The invention relates to the technical field of network flow processing, and discloses a cross-message HTTP domain name matching method and system on network shunting equipment, and the method comprises the steps: receiving and analyzing TCP segment messages, obtaining payload data and quintuple information, and executing the first domain name retrieval through a preset head anchor point and a preset tail anchor point; and if the head anchor point and the tail anchor point are simultaneously identified in the single message, directly extracting domain name information to complete matching, and if only one of the head anchor point and the tail anchor point is identified, marking the message as a to-be-recombined message and sending the message to a TCP segment recombining step. And when recombination is carried out, a recombination table item of the FPGA integrated HBM is accessed based on a quintuple hash value, payload data is written into a corresponding position, the recombination payload data is output when a preset output condition is met, finally, secondary domain name retrieval is carried out on the recombination data, and domain names are extracted to complete matching after head and tail anchor points are identified. According to the method, accurate matching of the cross-message domain names under the 400G large bandwidth is realized, the single-message line speed processing performance is considered, the problem of missed matching of the cross-message domain names is solved, and the method is adaptive to a network environment in which multiple websites share an IP (Internet Protocol).

The invention provides a semiconductor wire cutting tension adjusting device and method based on neural network optimization. The semiconductor wire cutting tension adjusting device comprises an algorithm processing central system, a parameter feedback system and a material cutting system. The algorithm processing central system comprises a system controller and initial parameter input; the parameter feedback system comprises linear cutting tension feedback, working temperature feedback and real-time cutting quality feedback; the material cutting system comprises a material platform, a cutting liquid conveying device, a roller structure, a take-up and pay-off wheel and the like. The method comprises the following steps: acquiring a large amount of parameter data such as wire aperture, material characteristics, cutting wire surface quality, wire tension, wire speed, material feeding rate, working temperature, cutting liquid flow, corresponding cutting rate, cutting precision and the like during different wire cutting operations in the early stage, pre-training through a neural network, and combining with a real-time optimization algorithm to obtain optimal cutting parameters, so as to achieve the purpose of improving the cutting efficiency. Therefore, higher cutting speed and higher cutting precision are obtained.

The invention discloses a machining process for cutting an 8-inch silicon rod by using a 70-micron-diameter diamond wire, which comprises the following steps of: filling a wire net with the 70-micron-diameter diamond wire in a double-roller multi-wire reciprocating cutting manner, putting a tool for bonding a crystal rod into an equipment worktable, locking, adjusting a cutting starting position to a position 1mm above the wire net, setting process parameters, and then cutting the silicon rod by using a diamond wire cutter, adding cutting fluid and then preheating and cutting; wherein the cutting tension is set to be 12-14 N, the acceleration time of the diamond wire is controlled within 5-6 s, and the wire consumption for machining each silicon wafer is 15-25 m; the cutting depth of the silicon rod ranges from 0 mm to 15 mm, the linear speed ranges from 1000 m / min to 1200 m / min, and the cutting speed ranges from 480 micrometers / min to 550 micrometers / min; the cutting depth of the silicon rod is 15-190 mm, the linear speed is 1200 m / min, and the cutting speed is 450-550 [mu] m / min; the cutting depth of the silicon rod is 190-210 mm, the linear speed is 1200-1000 m / min, and the cutting speed is 500-550 [mu] m / min. According to the invention, the silicon wafer with high-quality geometric parameters can be obtained.

This invention discloses a method for real-time monitoring and fine-grained correction of media data streams, relating to the fields of network security and streaming media processing technology. The invention achieves hardware-level line-speed data capture using XDP-eBPF-DPDK, generates robust intelligent fingerprints via a cross-modal graph attention network, employs PPO reinforcement learning for adaptive threshold decision-making, performs fine-grained transcoding correction using PTS / DTS calibration for abnormal content, and finally completes judicial-grade evidence preservation via a national cryptographic alliance blockchain. This invention supports 100Gbps line-speed processing, achieves a variant detection rate of 99.2%, and a false alarm rate of 0.5%, maintaining the continuity of media stream transmission while ensuring security, and is suitable for real-time compliance monitoring of ultra-high-definition advertising data streams.

The invention relates to the technical field of industrial robot network security, and discloses a robot cloud side-end collaborative security system based on a zero-trust architecture. The system comprises a cloud platform, an edge gateway and a robot controller, and is characterized in that a millisecond-level safety closed loop is realized through the following steps: a lightweight proving agent on the robot controller collects a system operation state and generates a proving voucher; calculating a behavior residual error between an actual moving track and an expected track of the robot by a digital twin service on the edge gateway; dynamically adjusting a certification strategy of the certification agent by a strategy engine on the edge gateway according to the behavior residual error; a capability token coupled with a robot motion control interpolation period is embedded into a data link layer message, and line speed verification is performed by using FPGA hardware, so that microsecond authorization and discarding of a data packet are realized. The method effectively overcomes the defect that a traditional safety scheme cannot complete real-time safety response in a robot control period.

The application discloses a kind of based on the quick flow identification and cache optimization method of heterogeneous network element equipment, it is related to the technical field of super high speed programmable data plane, wherein the method includes: original traffic is collected and feature extraction, constructs packet level and flow level feature set, and the structured flow feature vector that heterogeneous network element can be uniformly identified is compressed by deterministic integer operation generation;Based on the vector, construct extended flow state table, combined with native state table to establish two-level joint management mechanism, complete table entry insertion, update, aging and priority replacement;According to deterministic trigger rule, by programmable switching chip, flow key and feature vector are packaged and reported to field programmable gate array, realize remote state maintenance and result back, finally achieve high deterministic collaborative flow identification and cache optimization.The application realizes line speed forwarding and feature real-time synchronization by data plane deterministic pipeline processing, solves the problem of high flow identification delay and low efficiency.

The invention discloses a flow table state synchronization and packet processing method based on an intelligent network card, and belongs to the technical field of computer network communication, and the method comprises the steps: obtaining flow table item meta-information, local historical matching data and an external state prompt signal, and generating multi-dimensional prediction input; based on the multi-dimensional prediction input, a synchronization urgency index is obtained through calculation of a local flow table behavior predictor, and a precise synchronization request is generated and sent according to the synchronization urgency index; performing arbitration processing based on the local flow table mirror image and the shadow flow table area on the received data packet to generate a data packet processing result; and receiving pre-execution confirmation feedback generated based on the data packet processing result, and adjusting the calculation parameters of the local flow table behavior predictor according to the pre-execution confirmation feedback, so that the low-delay consistency synchronization of the flow table state and the continuous linear speed processing of the data packet can be realized while the control signaling overhead is remarkably reduced.

The invention provides a novel cast-rolling titanium wire speed monitoring visualization system, which comprises a laser engraving device used for periodically engraving a microgroove structure on the surface of a titanium wire, the depth of the microgroove being 20 [mu] m and the width being 50 [mu] m; the SAW resonator array emits surface acoustic waves to penetrate through the titanium steam layer; when the titanium wire moves, the surface microgrooves pass through a detection area of the SAW resonator array, the acoustic impedance characteristic is changed, and resonant frequency drift delta f is caused; establishing a mathematical model of frequency drift and speed: deltaf / f0 = k.v.cos theta, k being a material constant, v being a titanium wire speed, and theta being an included angle between the microgroove and a sound wave propagation direction; a time-frequency domain hybrid decoupling algorithm is adopted to eliminate roller vibration noise, the titanium wire movement speed is obtained through calculation, and high-precision, high-reliability and real-time measurement of the titanium wire speed is achieved.

The application discloses a kind of intelligent network card abnormal flow real-time monitoring and defense method and system based on light model, benchmark abnormal detection model is executed to forward load sensing compression, each layer of model is divided into micro-computing block according to computing amount, and the execution time of each micro-computing block does not exceed the preset quantile value of packet processing gap duration distribution, using hardware flow table matching engine, traffic is pre-screened into three levels of trusted release, high-risk block and detection, maintain inference state context for detection data stream, and execute micro-inference in packet processing gap stage by stage, trigger graded defense action according to abnormal probability score, drift detection is scheduled as low-priority microtask in gap and triggers incremental update when divergence value is out of limit.The application realizes abnormal flow real-time detection and active defense under the premise of guaranteeing line-speed forwarding.

The invention discloses an on-network aggregation method, device and system and a storage medium. The method comprises the following steps: S1, executing gradient aggregation logic by adopting a configurable matching table architecture; and S2, in the process of executing the gradient aggregation logic, Hash conflicts, resource overflow and floating point accumulation / sparse coding calculation are processed through an RISC-V calculation module. By adopting the technical scheme of the invention, a customized protocol and line speed processing specific to a data packet can be realized; and the abnormal condition which cannot be processed by the network card with limited network card flexibility can be processed.

The application discloses a high-speed serial connection drainage return flow abnormality detection message blocking implementation method based on FPGA, and the technical scheme process is as follows: receiving original message input FPGA, analyzing and extracting protocol fields, and then obtaining drainage message by selecting corresponding drainage strategy according to the fields; if there is no corresponding strategy, searching the high-speed flow table outside the FPGA and updating data, and the message hitting the drainage strategy is input into a processing strategy after mirror packaging for return flow and return staining judgment, and after success, the return flow message is unpacked and output, and the return staining message is stored in the high-speed flow table expansion data after flow table processing; the scheme adopts the collaborative idea of drainage and return flow, mirror and return staining, constructs a high-speed and efficient real-time network security protection system, relies on FPGA hardware acceleration to achieve line-speed processing, realizes high-density integration and low cost, can accurately detect and block network attacks, fills the gap, and promotes the independent controllability of network security core technology.

The application discloses a kind of multi-wire cutting prevent saw wire line arch excessive speed control method and system, this method is by establishing the indentation model of single diamond particle, deduces indentation depth and transverse crack length formula, obtains the indentation depth and transverse removal length of single diamond particle, and then calculates out single particle diamond abrasive removal material cross-sectional area;Again, the removal volume formula of macroscopic saw wire unit time is established in combination with abrasive density, thus the line arch stress simulation model is established, the relationship between line arch size and cutting force is clear, and critical line arch value is obtained;Line arch monitoring and speed adjustment are carried out to each speed interval of process in processing, and the control saw wire line arch is realized by the mutual cooperation of feed speed and wire speed.The method of the application can effectively avoid the continuous reduction of feed speed caused by detection, improve the problem of prolonged processing time caused by the reduction of feed speed, and improve the processing production efficiency.

The invention relates to a network message out-of-order damage simulation method, system and device and a medium, and relates to the technical field of network communication tests.The method comprises the steps that an out-of-order depth sequence of messages of the current batch is generated, and the sequence comprises at least one out-of-order depth value N; generating an out-of-order description array defining a message sending sequence by taking the out-of-order depth value N as array cache capacity; arranging the out-of-order description arrays corresponding to the out-of-order depth values according to a sequence in the out-of-order depth sequence, and constructing an out-of-order description sequence; performing hardware configuration based on the out-of-order depth sequence and the out-of-order description sequence; receiving and caching the messages according to the arrival sequence of the messages; and according to the currently configured disorder depth value N and the corresponding disorder description array, sending messages from the cache according to an array definition sequence. According to the invention, high-flexibility and configurable complex out-of-order scene simulation can be realized while the linear speed processing performance is ensured.

The application discloses a differential privacy big data processing method based on a programmable switch, wherein the method (1) utilizes a line-speed data packet processing ASIC pipeline to accelerate differential privacy calculation, thereby significantly reducing CPU cycle consumption; (2) reduces network flow overhead by aggregating network flow and removing unnecessary operations; and (3) automatically performs privacy auditing through the programmable switch. The application deploys a differential privacy system in a network, accelerates differential privacy calculation, improves CPU throughput, and reduces network flow and privacy auditing overhead.

This invention discloses an automated tantalum cleaning device and its cleaning control method, relating to the field of data analysis technology. It addresses the technical problem of dynamically adapting to load changes during the tantalum wire cleaning process to improve cleaning effectiveness and efficiency. The method includes: acquiring real-time parameters during the tantalum wire cleaning process; wherein the real-time parameters include the output current and amplitude values ​​of the ultrasonic generator, and the water resistivity value of the cleaning medium in the rinsing tank; determining the degree of detuning of the ultrasonic system based on the output current and amplitude values; determining the load level at the current cleaning position of the tantalum wire based on the degree of detuning, the output current and amplitude values; determining the ultrasonic power adjustment value and the wire speed adjustment value based on the load level and the water resistivity value; and controlling the cleaning device to clean the tantalum wire based on the ultrasonic power adjustment value and the wire speed adjustment value.