32results about How to "Reduce performance loss" patented technology

This invention relates to the field of computer technology, specifically to a method and apparatus for integrating third-party algorithm components based on dual-core collaboration. This application directly integrates third-party algorithm components into the BCM dual-core MCU, eliminating reliance on ZCU and CAN bus communication. This fundamentally eliminates the latency and security overhead caused by cross-ECU transmission and E2E protection, improving signal real-time performance and communication security. By clearly defining the functions and allocating resources of the BCM dual-core, a dedicated storage area is planned for the third-party algorithm components, and a standardized integration environment is built, achieving efficient and stable component integration. Simultaneously, peripheral access strategies are optimized, unnecessary protections and redundant polling are disabled, releasing idle computing power and hardware resources of the BCM and improving utilization. Without increasing the ZCU hardware cost, the chassis braking algorithm is deployed and executed locally, reducing performance losses caused by cross-domain communication and effectively improving the overall system operating efficiency and resource utilization efficiency.

The invention discloses a forming method of a magnesium oxide-based carbon dioxide adsorbent suitable for large-scale preparation. The forming method comprises the following steps: calcining a magnesium-containing precursor to prepare porous magnesium oxide raw powder; the preparation method comprises the following steps: carrying out impregnation modification on magnesium oxide raw powder and an alkali metal salt solution to obtain wet mixed thick slurry, drying the slurry, and carrying out secondary calcination to obtain modified adsorbent raw powder; and mixing the raw powder with deionized water, a forming aid and the like to prepare a plastic wet material, and performing extrusion-spheronization balling, drying and calcining to obtain regular spherical adsorbent particles. The formed adsorbent prepared through the method has the advantages of being controllable in particle size, high in mechanical strength, good in adsorptive property after being formed, excellent in cycling stability and the like, can be used for CO2 chemical adsorption trapping of CO2-containing gas under the medium-temperature condition, and is good in CO2 adsorptive property, economical efficiency of the technological method and environmental protection performance, the process technology is easy to amplify, and the method is suitable for industrial production. The method is suitable for industrial large-scale production.

This invention discloses a file decryption method, system, electronic device, and readable storage medium, particularly relating to the field of network security technology. The method includes: obtaining the address of a target encryption function in a process and instrumenting the target encryption function; monitoring file open operations and initializing the state record of the corresponding file in thread-local storage; intercepting the call to the target encryption function and reading the state record of the corresponding file in thread-local storage; if the state record meets the initial recording condition, extracting the encryption parameters and ciphertext fragments involved in the call to the target encryption function, storing the encryption event containing the encryption parameters and ciphertext fragments, and updating the state record; obtaining the file to be decrypted, extracting the data blocks of the file to be decrypted, and matching the data blocks with the ciphertext fragments in the stored encryption event; when a match is successful, performing a decryption operation on the file to be decrypted based on the encryption parameters in the matched encryption event.

The application provides a terminal and edge server cooperative model reasoning optimization method and system and a storage medium. The method comprises the following steps: acquiring the computing power parameters and communication parameters of local terminals and edge servers in an initial model deployment environment; combining the computing power parameters, the communication parameters and layer structure information of the initial model to calculate a total splitting cost; taking the total splitting cost and model reasoning accuracy as optimization objectives, processing the initial model based on an optimal pruning strategy to generate a lightweight model; selecting a target splitting point matched with a real-time communication rate from the matching relationship between the communication rate interval and the splitting point, splitting the lightweight model into a terminal-side sub-model and an edge-side sub-model based on the target splitting point, deploying the terminal-side sub-model on the local terminal and deploying the edge-side sub-model on the edge server to carry out cooperative reasoning. The embodiment of the application can ensure reasoning accuracy while realizing low delay.

The embodiment of the invention provides an Android cross-thread message tracking method, which belongs to the technical field of data processing, and specifically comprises the following steps: intercepting a native method related to message processing in an Android system; when a client thread delivers a Message object to the message queue, a recording process is triggered by an enqueueMessage method of Hook, and call stack information of the client thread is acquired and stored and is subjected to associated storage; when the server thread takes out the Message object from the message queue and processes the Message object, an application process is triggered through a DispatchMessage method of Hook, and the current server thread is associated with the Message object; in response to a request for obtaining the stack information, judging whether the current thread is associated with a message object which is being processed or not; through the scheme of the invention, efficient, stable and low-intrusion message tracking is realized.

The invention provides an image coding method and device, equipment and a storage medium, and relates to the technical field of computers. The method comprises the following steps that: a CPU (Central Processing Unit) analyzes an image to be coded in a wavelet transform coding process to obtain original pixel data, and transmits the original pixel data to a GPU (Graphics Processing Unit); the GPU performs preprocessing operation on the original pixel data in parallel by using a plurality of threads to obtain standard pixel data meeting the wavelet transform requirement; wherein one thread processes one pixel column in the original pixel data; the GPU performs forward wavelet transform on the standard pixel data to obtain a sub-band coefficient; and the GPU performs entropy coding processing on the sub-band coefficients to obtain coded data. According to the scheme, image analysis is completed in the CPU, and preprocessing, wavelet transform and entropy coding processing are executed in parallel in the GPU, so that the coding throughput rate and the resource utilization rate in the image coding process can be improved.

The application discloses a method for instruction branch prefetching in an AIGPU processor, which comprises the following steps: 1) defining a branch prefetching instruction in a GPU instruction set, the branch prefetching instruction is used to check whether the instruction at label exists in the cache, if the instruction at label does not exist in the cache, one or more instruction data at label are fetched from the lower storage to the cache; 2) according to the PC absolute value or relative value coded by the branch prefetching instruction, the data at the program branch instruction address is prefetched. The application provides a mechanism for branch prefetching through an instruction, which can reduce the performance loss during branch jumping and improve the system performance. The application does not use a complex branch prediction circuit, and the prediction delay and prediction error are avoided.

A beamforming design method for intelligent metasurface arrays based on orthogonal time-frequency-space technology includes: aggregating multipath channels under direct illumination into cascaded channels with near-field direction vectors; randomly initializing the phase shift of intelligent metasurface units; establishing an effective channel matrix based on the cascaded channels; then establishing a multi-user and rate objective function; maximizing the objective function through phase shift optimization using a hierarchical search strategy; repeatedly executing the hierarchical search strategy to search and optimize subsequent blocks until the number of intelligent metasurface units in each block reaches a preset value; then performing an exhaustive search in the final block, thus completing one iteration; repeating the iteration until the maximum number of iterations is reached; outputting the final optimal phase shift; and implementing beamforming design based on the final optimal phase shift. This invention can reduce the complexity of beamforming design and ensure reliable transmission of high capacity during high-speed movement.

This invention relates to the field of high-performance computing and provides a method and apparatus for optimizing the performance of a graphics processing unit (GPU) aggregate communication operator. The method includes: allocating device-side storage space on the host side; copying the structured data to be communicated to the device-side storage space to obtain a target data pointer; adjusting the kernel function parameter of the aggregate communication operator from a structure type to a pointer type and passing the target data pointer; executing the kernel function on the GPU side, directly accessing the device-side storage space based on the target data pointer to complete the aggregate communication operation; and releasing the device-side storage space after the aggregate communication operation is completed. This invention solves the problem of excessive GPU register pressure and decreased cache hit rate caused by inefficient memory access instructions generated by the compiler to pass structure parameters in existing technologies. It avoids the generation of inefficient instructions, reduces register pressure, and improves execution efficiency.

The invention provides a preparation method of a spherical padding for sintering a resistor disc of a lightning arrester, and belongs to the technical field of manufacturing of lightning arresters. The method comprises the following steps: firstly, preparing slurry by taking granulation waste as a raw material, and then preparing the spherical padding by virtue of a breakthrough titration forming process for the first time, so that the forming precision is high, the spherical quality is excellent, the use performance of the padding can be directly improved, the material structure is stable, the performance loss is low, and the original characteristics of the waste can be reserved to the greatest extent; the damage to the material performance in the preparation process is reduced. In addition, the preparation method provided by the invention is simple, the process is mature, and the use amount of additives is small, so that the production cost can be effectively reduced, and the economic value is relatively high.

The invention relates to a channel estimation method and device, communication equipment, a chip and a chip module. The method comprises the following steps: performing singular value decomposition on a time domain autocorrelation matrix and a frequency domain autocorrelation matrix of a target channel to obtain a time domain unitary matrix and a time domain eigenvalue corresponding to the time domain autocorrelation matrix and a frequency domain unitary matrix and a frequency domain eigenvalue corresponding to the frequency domain autocorrelation matrix; updating the time domain feature value based on the frequency domain feature value to obtain an updated time domain feature value; updating the frequency domain feature value based on the updated time domain feature value to obtain an updated frequency domain feature value; obtaining a time domain filtering coefficient based on the updated time domain characteristic value, and obtaining a frequency domain filtering coefficient based on the updated frequency domain characteristic value; and obtaining a filter coefficient based on the time domain filter coefficient and the frequency domain filter coefficient, and performing channel estimation based on the filter coefficient to obtain a channel estimation result of the target channel. By adopting the method, the channel estimation performance loss can be reduced.

The invention provides a neutron transport scanning operator optimization method based on built-in function and instruction level parallelism, and the method comprises the steps: firstly reconstructing a data layout, declaring an array according to an energy group sequence, carrying out the memory alignment, and enabling a calculation process to adapt to an SIMD vectorization architecture; then implementing explicit vectorization calculation by adopting a built-in function, performing polar angle cycle expansion and instruction recombination, and fully mining instruction-level parallel potential; meanwhile, key index calculation is optimized into vectoring polynomial approximation. Based on the optimized data flow and the parallel computing mode, efficient updating of the angular flux and the standard flux is achieved, and convergence judgment and calculation are completed. According to the method, the calculation throughput rate and the hardware resource utilization rate of the scanning operator are remarkably improved, and the method can be coordinated with coarse-grained parallel strategies such as MPI / OpenMP and is suitable for a multi-architecture high-performance calculation environment.

The invention belongs to the technical field of computer virtualization, and particularly discloses an elastic hierarchical scheduling method and device for graphics card virtualization, equipment and a storage medium. According to the method and the device, when the video card request command information of the virtual machine is received, the scheduling layer where the virtual machine is currently located is determined according to the analyzed virtual machine identifier; determining a target graphics card resource allocation strategy according to the scheduling layer; when it is determined that the cross-layer adjustment resource allocation condition is not met, graphics card resources dynamically scheduled to a scheduling layer are calculated; and distributing the video card resources according to the target video card resource distribution strategy. Through the mode, the performance loss of the virtualized graphic processing task is reduced by adopting the para-virtualization framework, when it is determined that the cross-layer adjustment resource allocation condition is not met, fair allocation and efficient utilization of graphics card resources are achieved according to the target graphics card resource allocation strategy, dynamic allocation of resources and adaptation of task loads in a multi-virtual-machine environment are guaranteed, and the resource allocation efficiency is improved. Therefore, the accuracy of dynamic resource allocation and the resource utilization rate can be effectively improved.

The invention discloses a robot slope control method and device, a robot and a storage medium, and relates to the technical field of robots. The robot slope control method comprises the steps that the body inclination angle of the robot is obtained; if the machine body inclination angle meets the slope compensation condition, the current expected speed of the robot is adjusted according to the machine body inclination angle, and the target expected speed is obtained; and controlling the robot to operate based on the target expected speed. According to the method, the machine body inclination angle of the robot can be obtained, if the machine body inclination angle meets the slope compensation condition, the expected speed of the robot is compensated and adjusted through the machine body inclination angle, and the robot is controlled to run based on the compensated target expected speed; the deviation between the actual speed and the expected speed of the robot when the gradient occurs is compensated, the problems of landslide, slope rushing, out-of-limit and the like are avoided, and the efficiency loss is reduced.

The application provides a method for selectively recovering metal elements in waste batteries by using a functional deep eutectic solvent, which comprises the following steps: mixing choline chloride, phosphoric acid and a choline hydroxide solution to obtain a deep eutectic solvent DES; pretreating a positive electrode of a waste lithium iron phosphate (LFP) battery and a waste lithium cobalt oxide (LCO) battery; and immersing the positive electrode in the deep eutectic solvent DES to make Li + enriched into the leaching solution, Fe is converted into FePO4 and left in the residue, and then cobalt is recovered by "ethanol-induced precipitation", lithium is recovered by "pH-controlled precipitation", and the DES is regenerated. 2+ This method realizes the closed loop of lithium iron phosphate and lithium cobalt oxide battery recycling in a green and economic way through the strategies of segmented targeted leaching and selective coordination precipitation, combined with solvent regeneration.

The application discloses a virtual scene-based data playback method, related equipment and a storage medium. The method comprises the following steps: acquiring a recording file of a virtual scene in response to a recording playback instruction; the movement event information in the recording file comprises position information of a virtual object in the virtual scene at a plurality of recording time points; playing back the virtual scene based on an interactive application, determining an associated recording time point closest to a current playback time point in the movement event information; determining target position information corresponding to the current playback time point based on position information of a virtual object corresponding to the associated recording time point; determining acceleration information based on the position information corresponding to the associated recording time point, the target position information and current position information corresponding to the current playback time point; and moving the virtual object in the virtual scene to the target position based on the acceleration information. The application reduces the performance loss of the client in the playback process, and the playback movement is closer to the movement of a real object.

The invention discloses a disk capacity management method and device for a database cluster, equipment and a medium. Comprising the following steps: when a new node is allocated to a target fragment, performing directory storage quota configuration on a disk mounted in the new node according to a disk quota template of the target fragment; wherein the disk limit template comprises a directory type, a capacity limit type, a capacity limit value, an alarm threshold and a full strategy; regularly collecting disk storage information in the target node, and calculating the disk utilization rate of the first directory according to the collected current disk capacity, the used capacity of the first directory and the capacity limit value of the first directory; and comparing the disk usage rate of the first directory with the alarm threshold value of the first directory, and performing access management on the target node when determining that the disk usage rate is higher than the alarm threshold value of the first directory or the usage rate is full. By adopting the technical scheme, the directory use capacity can be monitored and dynamically managed based on the multi-stage disk quota configuration.

The application relates to a charging method and device, computer equipment, a storage medium and a computer program product. The method comprises the following steps: after determining that a charging interface of a backup battery establishes a connection, determining whether the backup battery can be charged according to a vehicle initial ignition state, an initial temperature of the backup battery and a pre-configured temperature condition range; in the case of determining that the backup battery can be charged, acquiring a voltage of the backup battery and a duration of the vehicle ignition state; determining a charging mode of the backup battery according to the voltage of the backup battery, a pre-configured voltage condition threshold and the duration of the vehicle ignition state, and controlling the charging of the backup battery according to the charging mode, wherein the charging mode comprises a constant current mode or a trickle current mode; and determining whether to stop charging the backup battery according to the charging mode and the pre-configured temperature condition range. The method can reduce the performance loss of the backup battery and is beneficial to improving the service life of the backup battery.

The present disclosure provides a kind of SFTP connection leakage detection method, system and device.The system includes intelligent access module, creates session monitoring context when establishing SFTP connection, including session identification, operating state flow vector and current thread stack snapshot;Traffic probe module obtains connection real-time byte throughput information;State update module updates the current state of operating state flow vector, history list and cumulative transmission byte number based on the information;Analysis and reporting module calculates dynamic baseline speed according to historical sampling points and determines dynamic decision threshold, determines whether there is leakage by comparing current transmission speed with threshold value;If there is leakage, locate fault location based on session identification and stack snapshot and generate exception report.

The invention discloses a method, a system, equipment and a medium for recording an original voice stream in real time and converting the original voice stream into a playable voice file, and the method comprises the following steps: after a calling party and a called party establish a call, combining SDPs of the calling party and the called party into the same SDP capable of distinguishing the calling party and the called party, sending an SIP signaling to a naked stream recording server through an SIP protocol, after a recording request is received, a calling SDP and a called SDP in the SIP signaling are analyzed, a media format is obtained through media level description and fmtp in the SDP, and received voice media streams are classified and stored according to the media format; the media format is divided into constant packet media and non-constant packet media; for the non-constant packet media, a mute packet is supplemented to the voice media stream; respectively compressing and packaging the constant packet media and the non-constant packet media supplemented with the mute packets to obtain a playable audio file; and combining the playable audio files of the calling party and the called party into a dual-track audio file. According to the invention, the system performance consumption caused by transcoding and frame supplementation required by original recording is reduced, and the concurrency performance of stand-alone voice recording is improved.

This invention discloses a construction process for mining anchor bolts, relating to the field of mining anchor bolt construction technology, including the following steps: raw material preparation and pretreatment; preparation of functionalized rolling penetrant; heating and roughing / intermediate rolling; online reactive finishing rolling; cooling and shaping; and post-treatment. This invention constructs a gradient composite modified layer on the anchor bolt surface through online reactive rolling, achieving interfacial chemical bonding and grain boundary indentation of nano-functional units. This not only reduces installation frictional resistance but also enhances the material's corrosion resistance and hydrogen embrittlement resistance through a dense shielding network and self-healing mechanism, ensuring the reliability of support in complex environments.

The invention belongs to the technical field of quadrotor unmanned aerial vehicles, and particularly relates to a quadrotor unmanned aerial vehicle mapping and planning method based on deep reinforcement learning, and the method comprises the following steps: S1, feature extraction and screening, S2, feature map creation, S3, feature map position calculation based on value iteration, and iteration through geometric verification and using a TD algorithm, and S4, state estimation. A planning method of quadrotor unmanned aerial vehicle mapping based on deep reinforcement learning is used for planning a generated flight trajectory of quadrotor unmanned aerial vehicle mapping, and the planning method comprises the following steps: S5, generating a strategy network model of the trajectory; according to the method, the common feature points and the feature point set between the frames can be rapidly extracted, the correlation graph of the feature points and the feature set is determined, and finally the corresponding relation and the position of the graph are calculated through a reinforcement learning method to obtain the accurate feature point relation.

The invention relates to a fairness repairing method for constructing a table data classification model of multiple loss items, and belongs to the field of artificial intelligence and machine learning. The method comprises the following steps: firstly, counting the number proportion of various different sensitive attributes in a training set, adjusting the weight of a fair regular term of a loss function according to the proportion to retrain a model, and calculating a change ratio of a fairness index to an accuracy index by utilizing a test set; secondly, randomly replacing each attribute of the table with a reconstructed sample, and calculating the similarity between an original sample and the reconstructed sample as a model comparison loss item; and finally, adjusting the weight of a fair regular term and comparing the weight of a loss term by using the change ratio, and retraining the model to generate a fair repair model. Aiming at the problems that prediction fairness imbalance of a model to other sensitive attributes is easily caused by only processing a single sensitive attribute and the accuracy is influenced by deleting the sensitive attributes in the existing method, fairness repair under the condition that the model accuracy is kept is realized through training sample reconstruction and sensitive attribute weight distribution.

The invention relates to the technical field of chemical equipment, and discloses a quick assembly device and a quick assembly method of a modular ammonia still tray and an ammonia still system. The quick assembly device comprises a tower body, a maintenance channel, a treatment cover and a push-out mechanism. A tower tray capable of horizontally moving is arranged in the tower body, a maintenance channel with a detachable cover is arranged on the side wall of the tower body, the treatment cover is fixed on the outer wall of the tower body and is communicated with the maintenance channel, and a treatment cavity is formed in the treatment cover; the push-out mechanism is located on the side, opposite to the maintenance channel, of the tower tray in the tower and comprises a pushing block and a driving piece. And when the opening cover is opened, the driving piece drives the material pushing block to push the tray into the processing chamber through the maintenance channel. According to the device, the tray is laterally moved out, so that non-entering type rapid maintenance is realized, the production interruption time is reduced, the safety risk that personnel enter a limited space for operation is avoided, and the maintenance labor intensity is reduced.

The utility model relates to the technical field of bearings, in particular to an axial-radial large-load ball bearing which comprises an outer ring, an outer check ring, a retainer and an inner ring, an outer raceway is arranged on the inner circumferential surface of the outer ring; the outer check ring is operably associated with the outer ring; the retainer is provided with at least ten rows of pocket hole groups, pocket holes of each row of pocket hole groups are distributed in an annular array around the axis of the retainer, and a steel ball is arranged in each pocket hole; the low-speed heavy-load ball bearing is suitable for low-speed heavy loads, the axial bearing capacity of the ball bearing is improved, the requirement for bearing movement is met during radial rotation, and when the inner ring and the outer ring are aligned, the contact point of the steel ball on the outermost side and the inner ring can move to the linear ending position of the inner ring on the side corresponding to the contact point of the steel ball on the outermost side.

This invention provides a method, system, electronic device, and storage medium for secure management and rapid revocation of P2P access to CXL devices. The method includes: a management entity registering the device and pre-setting keys, policies, and revocation version numbers; the access initiating device generating a security token containing a chained integrity tag based on the policy and sending it to the accessed device; the accessed device centrally verifying the token through multi-level pipelined verification, and establishing an active session upon successful verification; during data plane access, the request only requires a session identifier and a short authentication code, which the accessed device can execute through a quick table lookup verification; the management entity monitors error statistics and triggers version number updates, thereby quickly rejecting all accesses associated with the old version by the accessed device. This invention achieves hardware-based, low-latency end-to-end secure authorization, supports fine-grained permission tightening delegation, and enables immediate revocation of global access credentials through a versioning mechanism, significantly improving the security and performance of CXL P2P access scenarios.

The embodiment of the application discloses a data processing method of a shared memory, related equipment, products and media, which can be applied to the technical field of data processing. The method comprises the following steps: receiving a data processing request initiated based on a computing task of a distributed application, and detecting a request result of the data processing request in a local memory; a plurality of devices are used for processing the computing task of the distributed application; if the request result of the data processing request is not detected in the local memory, the request result of the data processing request is detected in a shared memory; if the request result of the data processing request is not detected in the shared memory, the request result of the data processing request is calculated in the local memory, and the calculated request result is added to the shared memory. By adopting the embodiment of the application, the processing efficiency of the computing task of the distributed application can be improved. The embodiment of the application can also be applied to various scenes such as cloud technology, artificial intelligence, intelligent transportation and auxiliary driving.

This invention relates to a method, apparatus, and storage medium for black-box code analysis and test case generation, applied in the field of software testing technology. It includes: non-intrusive acquisition of variable change trajectories and execution paths within black-box code, making previously invisible internal logic states transparent, enabling testers to accurately locate intermediate logic errors, and solving the problem of traditional black-box testing's inability to observe internal states. Secondly, by systematically reconstructing the logical structure and execution path of the code, this application can accurately identify uncovered logical branches and intelligently reverse-engineer and generate supplementary test cases, overcoming the shortcomings of traditional black-box testing, such as reliance on experience and incomplete coverage, significantly improving the depth and efficiency of testing. Thirdly, this application uses a system tracing interface for monitoring, without modifying the target code's bytecode, avoiding the potential impact of intrusive methods on program performance and stability.

The invention provides an online auction information processing system and method supporting bidding of multiple buyers and sellers, and relates to the technical field of electronic commerce. The online auction information processing system and method supporting the bidding of the multiple buyers and sellers comprises the steps that S1, bidding requests of the multiple buyers and sellers are received based on an event-driven network communication mechanism, monotonically increasing logic time sequence identifiers are distributed to each bidding request, and bidding data are written into a memory type ordered queue, and forming a bidding event stream arranged according to a logic time sequence. A multi-source bidding request is received in an event-driven network communication mode, and a monotonically increasing logic time sequence identifier is introduced for each bidding behavior, so that the bidding behaviors form a traceable and comparable logic sequence in a concurrent environment, and interference on matching judgment caused by real physical time jitter is avoided.