5945 results about "Parallel processing" patented technology

The invention relates to the technical field of data processing, in particular to a storage and calculation integrated parallel processing system and method. The method comprises the following steps of obtaining original stored data and performing topological skeleton projection, generating a projection skeleton structure to realize characteristic hierarchical integration, determining a weight relation matrix based on multi-scale space fusion data, realizing matching of a memory and a calculation unit through gradient optimization adjustment, and obtaining a multi-scale space fusion matrix. And the mining storage and calculation unit activates the chain type interaction relationship of the mapping data and performs recursive self-calibration, generates self-calibration update data, predicts a heterogeneous access channel and reconstructs a data access link, and finally performs task recombination and pipeline scheduling and eliminates parallel conflicts, thereby obtaining an efficient parallel processing result. According to the method, high efficiency and intelligentization of storage and calculation are realized, and reliable support is provided for execution of complex calculation tasks.

The invention relates to the technical field of radar control, in particular to an anti-interference control method applied to a radar, and aims to solve the problem that an existing radar anti-interference technology is insufficient in anti-interference capability in a complex electromagnetic environment. According to the method, multi-dimensional input data of a radar system are synchronously collected, interference feature extraction is performed on multi-dimensional data streams by using a dual-channel deep network, and interference fingerprint vectors are output to a dynamic feature library in real time. And based on the interference feature priority list, performing cooperative processing, including spatial filtering, frequency domain notch, polarization projection and residual extraction and sparse reconstruction of a target signal. And dynamically adjusting the processing parameters according to the real-time evaluation result. And data distribution processing is carried out on the suppressing interference and the deception interference, and final target trajectory information is output through result fusion. Parallel processing is also realized on the FPGA chip, and the processing speed is improved. On the whole, the anti-interference capability, the stability and the real-time performance of the radar system in a complex electromagnetic environment are improved.

The invention discloses a big data distributed storage and parallel processing collaboration method based on cloud computing. The method comprises the following steps: sensing data stream characteristics in real time through a self-adaptive dynamic partitioning engine, dynamically adjusting a partitioning strategy and generating a metadata label; constructing a node selection model through a comprehensive evaluation algorithm, selecting storage nodes to form an optimal storage cluster, and dynamically adjusting a resource matching weight coefficient based on a load state through a load optimization module; decomposing a data processing task into parallel subtask units, and constructing a dual-objective optimization model; triggering a dynamic rebalance mechanism through a distributed monitoring agent in combination with a hierarchical early warning strategy; and constructing a multi-level cache system to optimize a data access path, outputting a final result, pushing the final result to the user terminal, and updating the knowledge base. Through collaborative optimization of dynamic partitioning, multi-dimensional resource scheduling, elastic scaling and intelligent caching technologies, the resource utilization rate, the load balancing capacity and the stability in a high-concurrency scene of the system are remarkably improved.

Techniques are described for partitioning B-tree keys of file systems into key ranges for parallel processing in delta generation during file storage replications between file systems in different cloud infrastructure regions. In certain embodiments, a delta generation processing for cross-region replication may utilize a key-range splitting mechanism involving a recursive algorithm that partitions B-tree keys of a source file system into roughly equal-size key ranges. All the partitioned key ranges may be processed in parallel and concurrently by different processing threads, one thread per key range, to improve the performance of the delta generation and achieve scalability.

The invention provides a supply chain collaborative material management system and method, and relates to the technical field of material management, and the method comprises the steps: receiving a multi-enterprise heterogeneous material data flow, and constructing an industry knowledge graph based on the field-level semantic mapping of a dynamic ontology library; a standardized data table is generated; standardized data is injected into a distributed message queue for parallel processing, dynamic cache partitions are created according to material object labels, timestamp indexes are built in the dynamic cache partitions, a sliding window mechanism is adopted to scan data, and multi-strategy resolution is performed on conflict data in a window; performing conjoint analysis on the cleaned data to generate an enhanced analysis report, and activating a case transfer learning mechanism to update a model when the similarity between new data and historical cases exceeds a threshold value; constructing a material state transfer matrix for the abnormal data, simulating a correction operation in a digital twin environment, and triggering automatic correction; the problems of multi-enterprise heterogeneous data integration, dynamic data processing, intelligent analysis and the like are solved, and the collaboration efficiency and the management level of a supply chain are improved.

The invention discloses a self-adaptive packet data transmission system and method based on Beidou short messages, the system is composed of eight core modules such as a serial port communication module, a protocol conversion module and a configuration module, multi-link communication is established through an RS232 serial port and a Beidou director, and Baud rate self-adaptive adjustment and multi-serial port parallel processing are supported. The protocol conversion module adopts a finite-state machine to analyze a BDS-3 protocol, data fragmentation and recombination are achieved in combination with a sliding window algorithm, and the compression rate reaches 3: 1. The method comprises the following seven stages of flows: evaluating channel quality by detecting a message, and dynamically switching a BDS-3 anti-interference mode; implementing flow control by adopting a token bucket algorithm, and dividing three levels of QoS guarantee emergency instructions; and terminal health management is realized through a heartbeat detection and diagnosis instruction set. According to the method, instruction execution is delayed by 1.2 seconds in emergency communication, the bit error rate is controlled to be 0.0005%, the transmission efficiency is improved by 40% compared with a traditional scheme, and the method can be widely applied to the fields of Internet of Things, intelligent transportation and the like.

The invention relates to the technical field of dam deformation monitoring, and provides a dam deformation monitoring and early warning method and system based on Beidou and machine vision, and the method comprises the steps: collecting the original displacement data of a dam and the original image data of the surface of the dam; performing RTK differential processing and multi-point positioning data weighted average on the original displacement data to obtain dam displacement data, and performing multispectral analysis and deep learning processing on the original image data to obtain dam surface structure feature data; carrying out localization processing on the dam displacement data to obtain real-time displacement change data, and carrying out parallel processing on the dam surface structure feature data to obtain real-time deformation feature data; dynamically fusing the real-time displacement change data and the real-time deformation characteristic data to obtain comprehensive deformation data; and establishing a dam deformation trend prediction model according to the comprehensive deformation data, and pushing graded early warning information to a monitoring center when abnormal deformation is detected. According to the invention, the real-time performance of dam deformation monitoring and the early warning reliability are improved.

The invention discloses a tensor core-based method for accelerating arbitrary precision sparse matrix multiplication and addition operation, which comprises the following steps of: aiming at an architecture of a tensor core on A100 and a programming model thereof, combining program characteristics of deep learning parallel processing sparse matrix, combining an arbitrary precision sparse matrix multiplication and addition operation operator with parallel computing acceleration equipment, and performing parallel computing acceleration on the sparse matrix. The method comprises the following steps: decomposing a general matrix into a sparse matrix in a set format, partitioning the sparse matrix and assigning a corresponding parallel computing hardware unit for the sparse matrix, reducing memory conflicts and access delay by using a shared memory, calling a high-level language API function library to execute loading, computing and storing processes of a computing data volume, and finally performing specification on output data of a function. Therefore, the final result of the random-precision sparse matrix multiplication and addition operation on the parallel computing acceleration equipment is obtained. According to the method, parallel computing resources such as the tensor core on the A100 can be fully utilized, the program calculation amount and the communication overhead are effectively reduced, and the program operation performance is greatly improved.

The invention provides an intelligent agent automatic generation and scheduling system based on an artificial intelligence large language model. The system comprises an intelligent agent configuration module driven by the large language model, a workflow engine and automation module and a front-end and rear-end integration module. The agent configuration module driven by the large language model realizes automatic configuration of an agent and a workflow based on the large language model, and comprises a large language model interaction layer and an agent configuration module; the workflow engine and automation module is responsible for execution logic of workflow, supports multi-task parallel processing and comprises a workflow engine module and a task scheduler; the front-end and rear-end integration module realizes cooperative work of a JavaScript front end and a Python Django rear end by designing a unified data interface, the front end is responsible for interacting with a user and displaying an execution state of a workflow, and the rear end provides service logic processing and data persistence.

The invention relates to the technical field of cloud computing, and discloses a resource scheduling method, device and equipment, and the method comprises the steps: generating a multi-dimensional data topological graph of cloud environment monitoring data, decomposing a to-be-scheduled job set into quantum computing unit sets which can be processed in parallel, and adding priority labels for the quantum computing unit sets through a quantum heuristic algorithm, according to priority labels and a multi-dimensional data topological graph, physical resources of a cloud environment are divided into non-uniform resource fragments matched with task requirements, based on the non-uniform resource fragments, resource scheduling is carried out through an improved hybrid particle swarm optimization algorithm, and weighting and fitness evaluation and resource constraint correction are introduced. According to the resource scheduling method and device, the multi-dimensional data topological graph is generated to comprehensively present the resource state and association, then the to-be-scheduled job set is decomposed into the quantum computing unit sets, the priority labels are added, and then the non-uniform resource fragments are divided for resource scheduling, so that the resource adaptation accuracy is improved, and the scheduling efficiency and the resource utilization rate are improved.

The invention relates to the technical field of hard disk testing and verification, in particular to a multi-thread-based solid state disk reliability automatic testing method and system.The method comprises the steps that firstly, SMART information is deeply analyzed through microsecond-level high-granularity continuous performance monitoring, and multi-thread parallel processing is assisted; according to the method, fine performance fluctuation of the solid state disk under the concurrent load can be quickly captured, a fault mode can be identified, then early warning is realized by utilizing the extracted multi-dimensional features and a machine learning model, and a detailed fault diagnosis report is generated; and through dynamic error correction code strength verification and data integrity verification under pressure, an internal error correction mechanism of the solid state disk is actively detected and optimized. And finally, in combination with prediction reliability modeling, the system can estimate the remaining service life and predict faults, and provides product optimization suggestions for design, manufacturing and firmware optimization of the solid state disk, so that automation, intelligence and full life cycle management of the fault detection reliability of the solid state disk are realized.

The invention relates to an audio noise reduction method, device and system based on deep learning, and the method comprises the steps: obtaining an input audio signal with noise, and carrying out the multi-scale time-frequency decomposition, and obtaining a mixed time-frequency feature and a noise fingerprint spectrum; performing parameter parallel processing on the noise fingerprint spectrum through a preset dynamic kernel generation network, and performing preliminary noise reduction processing on the mixed time-frequency characteristics to obtain noise-reduced mixed data; performing dual-path processing structure construction on the noise reduction mixed data to obtain amplitude optimization data and phase optimization data; performing dynamic time-frequency domain cross fusion on the amplitude optimization data and the phase optimization data to obtain fused audio data; and carrying out differentiable acoustic equation constraint adversarial training on the fused audio data, and carrying out inverse time-frequency transformation processing to obtain a target noise-reduced audio signal. According to the invention, the overall efficiency and effect of audio signal processing can be effectively improved.

The invention relates to the technical field of medical systems, and particularly discloses an anesthesia complication prediction model construction method based on deep learning, and the method comprises the following steps: S1, obtaining multi-source heterogeneous anesthesia medical data; s2, constructing a multi-modal feature fusion module; s3, designing a hierarchical deep neural network architecture which comprises sub-networks for processing different modal data in parallel and a full-connection prediction layer fusing multi-modal features; s4, continuously outputting a complication probability curve in a sliding time window mode by adopting a dynamic risk trajectory prediction mechanism instead of a single static prediction result; and S5, deploying a clinical real-time decision interface, and mapping a prediction result to an anesthesia monitoring equipment alarm system in real time. A bidirectional LSTM + 1D-CNN hybrid encoder and a cross-modal attention mechanism are adopted, time sequence dependence of physiological signals and spatio-temporal characteristics of operation events are synchronously captured, deep semantic fusion of multi-source data is achieved, and the characterization capacity of a model for precursor characteristics of complications is improved.

The invention discloses a PCIe switch multicast processing system and method giving consideration to efficiency and resources, and the system comprises a unicast control module which is used for receiving the target port information of a unicast TLP packet, a TLP indication signal, TLP header information and TLP load data, and transmitting a unicast request; the multicast control module is used for receiving a target port list of a multicast TLP packet, a TLP indication signal, TLP header information and TLP load data, and sending a multicast request; the target port arbitration module is used for receiving a unicast request and a multicast request, and sending a request packet corresponding to the unicast request or the multicast request to the sending cache module; and the sending cache module is used for receiving the authorized unicast TLP packet or multicast TLP packet, caching the unicast TLP packet or multicast TLP packet, and sending and outputting the unicast TLP packet or multicast TLP packet to the egress end of the target port. According to the invention, the processing of the multicast packet is centralized at the ingress end of the PCIe port, the unicast packet and the multicast packet are separately processed, meanwhile, two parallel processing channels are opened up in the multicast control module, and the copying of the multicast packet is realized through a roll back FIFO read-write pointer mode, so that the design is simplified, the resources are saved, and the efficiency is improved.

A parallel processing unit comprises a plurality of processors each being coupled to a memory access hardware circuitry. Each memory access hardware circuitry is configured to receive, from the coupled processor, a memory access request specifying a coordinate of a multidimensional data structure, wherein the memory access hardware circuit is one of a plurality of memory access circuitry each coupled to a respective one of the processors; and, in response to the memory access request, translate the coordinate of the multidimensional data structure into plural memory addresses for the multidimensional data structure and using the plural memory addresses, asynchronously transfer at least a portion of the multidimensional data structure for processing by at least the coupled processor. The memory locations may be in the shared memory of the coupled processor and / or an external memory.

The invention discloses a railway office system heterogeneous database integration method based on a data conversion technology, and relates to the technical field of data integration, the method realizes seamless integration of heterogeneous databases in a railway office system by uniformly designing a route data standard model and combining an AI intelligent mapping algorithm; according to the scheme, a distributed architecture is adopted to process data conversion tasks in parallel, node loads, conversion speeds and parallel speed-up ratios are monitored dynamically, and the system stability in a high-load scene is ensured; the universal adaptation layer supports integration of multiple protocols and multiple interfaces, and the overall performance is effectively improved; meanwhile, a multi-level security mechanism and consistency verification are introduced in the scheme, dynamic encryption, authority management and data redundancy backup are combined, the integrity and security of data circulation are guaranteed, and the problems of complex data integration, poor compatibility and performance bottleneck are solved.

The invention discloses a sampling implementation method and system suitable for motor protection quick response, belongs to the technical field of motor protection, and can increase the sampling frequency and improve the quick response of motor fault protection. Comprising the steps that motor phase current and bus voltage are collected, after anti-aliasing filtering is conducted, parallel sampling is conducted through a main ADC module and a redundant ADC module which are independently configured, and the following operations are synchronously executed based on a double-buffer-area framework: in a first buffer area, a compensation coefficient updating period is dynamically adjusted, and convergence is restrained; in the second buffer area, generating a voltage gradient prediction sequence, and compensating data abrupt change; when the harmonic frequency spectrum amplitude or the voltage gradient prediction sequence exceeds the limit, an oversampling mode is triggered, a redundant ADC module is allocated to load a multi-order digital filter, and an independent hardware acceleration unit is started to process harmonic compensation and gradient prediction in parallel; and generating a dynamic safety envelope, and if the gradient extreme value in the preset continuous window breaks through the safety envelope, generating an overvoltage fault signal.

The invention relates to the technical field of Bluetooth earphones, and discloses a Bluetooth earphone automatic pairing method and device for multi-terminal switchover, and the method comprises the steps: carrying out the equipment information scanning of a preset region range of a Bluetooth earphone, obtaining an equipment identifier, a signal intensity value and an equipment state parameter, and storing the equipment identifier, the signal intensity value and the equipment state parameter in a priority database; inputting a double-layer deep belief network to perform feature mapping calculation to obtain an equipment connection score; establishing a first equipment priority queue and performing dynamic updating calculation to obtain a second equipment priority queue; distributing independent signal processing channels for parallel processing to obtain effective connection equipment; performing periodic acquisition, and generating a switching instruction when a difference value exceeds a preset threshold value; according to the method and the device, the probability of switching failure is effectively reduced, and the stability of automatic pairing of the multi-terminal switching Bluetooth headset is improved.

The invention provides a deep reinforcement learning optimization algorithm based on an adaptive encryption framework, and the algorithm comprises ten steps: data input preprocessing, data sensitivity evaluation, system resource state acquisition, state vector construction, strategy decision, encryption execution, calculation reward, model updating, result output, and loop optimization. The method aims at solving the problems that encrypted data processing efficiency is low and safety is difficult to consider. According to the method, intelligent optimization of the encryption strategy is realized through the data sensitivity evaluation module, the adaptive encryption strategy adjustment module, the deep reinforcement learning optimization module, the parallel processing module, the resource scheduling module and the like. On the premise that the data security is ensured, the processing efficiency of the system is remarkably improved, and the method is suitable for scenes such as cloud computing, big data processing and the Internet of Things which need to process a large amount of sensitive data.

The invention relates to the technical field of artificial intelligence, and provides an operator parallel processing method, electronic equipment, a storage medium and a program product, and the method comprises the steps: determining a step of data dependence based on each processing step of a first operator and a second operator; and performing parallel execution on each processing step of the first operator and the second operator by applying a synchronization strategy and a segmentation strategy based on the plurality of cooperative thread bundles. According to the method, the parallel fusion processing of operators is realized through the steps of determining data dependence, segmenting the data, realizing parallel execution by utilizing the cooperative thread bundle, inserting synchronization between the data dependence steps and the like, the calculation performance is remarkably improved, and meanwhile, the data blocks after segmentation are multiplexed, so that the data processing efficiency is improved. The data dependence in the operators and between the operators can be reduced, and the overhead of data reading, instruction synchronization and the like can be reduced.

Techniques are described for a hierarchical caching mechanism enabling efficient cross-region replications. In some embodiments, replication-related information (e.g., key-value pairs) is stored in a particular layout in a binary tree (B-tree) of a file system for replication processing. A hierarchy of caches storing a first type of information (e.g., crypto keys associated with iNodes) may be arranged to match the particular layout in the B-tree to enable efficient parallel processing of a second type of information (e.g., files, file data, or symbolic links), where the replication-related information in the B-tree is partitioned into multiple key ranges for parallel processing. In some embodiments, the caches in different hierarchies may be shared by different parallel-processing key ranges and replication jobs in a file system.

The invention provides a power grid energy disturbance suppression method for electric heating gas state information transmission collaboration, and relates to the technical field of information, and the method comprises the steps: obtaining multi-source heterogeneous state information of an electric heating gas energy network; aiming at the multi-source heterogeneous state information, dynamically converting a heterogeneous communication protocol into a unified protocol format by adopting a self-adaptive communication protocol conversion technology; and for the decoupled state information, parallel processing is carried out by adopting a distributed stream computing framework, so that the real-time performance and expandability of state information processing are improved. A distributed stream computing framework is adopted to construct an efficient processing assembly line, the block chain technology is utilized to guarantee the integrity and reliability of data transmission, a deep learning algorithm is applied to construct an intelligent analysis model, a multi-energy-flow coupling relationship and an evolution law are mined, and a multi-energy-flow analysis result is obtained. According to the method, data processing results of all links are fused to form a global state information view, decision support is provided for power grid energy disturbance suppression, and efficient processing, analysis and visualization of the state information of the electric heating gas multi-energy flow system are achieved.

Apparatuses, systems, and techniques to perform a neural network. In at least one embodiment, an application programming interface schedules two or more graph nodes to be performed by two or more parallel processing pipelines based, at least in part, on an order of layers in the neural network.

The invention provides a multi-transaction deadlock detection and processing method in resource-constrained equipment, which comprises the following steps of: in a transaction execution process, constructing a resource waiting graph which is used for recording a waiting relationship among transactions so as to reflect the occupation condition of shared resources by the transactions and a dependency relationship among the transactions; before a transaction is subjected to read-write operation each time, whether execution of the current transaction can cause a deadlock risk or not is judged in real time by detecting a dependency relationship between the current transaction and a target resource and combining a resource waiting graph; and when a deadlock risk is detected, deadlock avoidance strategies of priority processing, transaction rollback, transaction retry and exception processing are adopted to avoid deadlock. The deadlock problem occurring in the transaction parallel processing process can be effectively avoided or solved, and meanwhile the response efficiency of the application and the utilization rate of the storage space are guaranteed.

The invention discloses an audio stream adaptive decryption method, system and device and a storage medium. The method comprises the following steps: acquiring an MPEG-DASH list file and analyzing the MPEG-DASH list file; the standard DRM interface is initialized; performing feature extraction on the identified audio segmentation information in the audio segmentation mode, and dynamically adjusting a segmentation request and processing logic according to the extracted features to perform adaptive segmentation downloading and preprocessing; extracting the encryption parameter to identify an encryption mark of the target audio stream platform, and selecting a corresponding decryption algorithm according to the encryption mark to decrypt the current audio segment; and performing audio frame reconstruction on the decrypted data to form an updated audio stream, caching the updated audio stream to a preset cache region based on a self-adaptive buffer strategy, and outputting the final audio stream to ensure the continuity of the final audio stream and realize efficient parallel processing of decryption and playing, so that the decryption efficiency is improved, and the user experience is improved. And particularly, in a high-concurrency scene, the user experience is improved.

The invention relates to the technical field of distributed learning, in particular to a large model pre-training system based on distributed parallel processing. In the system, a data distribution layer collects a node resource state through a fragmentation module and generates a dynamic scheduling strategy; a computing resource layer configuration model initialization module and a strategy switching module support flexible switching of multiple modes such as tensor parallelism, data parallelism and assembly line parallelism; the network communication management layer is combined with topology perception and gradient compression technologies, so that the communication efficiency is improved; and the model aggregation layer realizes global parameter updating and training tuning under privacy protection through a security aggregation and optimization control mechanism. All layers of the system operate cooperatively, the calculation efficiency, the communication performance and the data security of large model pre-training can be effectively improved, and the method is suitable for model development and deployment in a large-scale heterogeneous calculation environment.

The invention relates to the field of power data management, and discloses a power big data adaptive management method and system fused with spatial-temporal feature mapping, and the method comprises the steps: obtaining a real-time operation data flow from a multi-source power terminal, and constructing an original power data set with a time sequence label and a device identifier; the method comprises the following steps: dividing an original power data set into parallel processing units based on a storage-while-computing architecture, performing dynamic index updating by adopting an event-driven index mapping rule, and constructing a multi-dimensional data index cache system with real-time responsiveness; identifying a key abnormal trajectory through a time-varying feature nesting mechanism, and performing hierarchical measurement and entropy disturbance analysis on a data fluctuation degree in the key abnormal trajectory by using a streaming feature aggregation network; identifying potential security risk nodes in combination with the structure matching degree between the historical abnormal event evolution graph and the key abnormal trajectory; and generating a multi-level response instruction chain based on the risk assessment result. The method has the advantage of improving the operation safety of the power grid.

The invention discloses an industrial image anomaly detection method based on deep learning, and particularly relates to the technical field of industrial visual detection. The problems of high false alarm rate, fuzzy fine defect positioning, insufficient real-time response capability, difficulty in model increment updating and the like caused by data distribution drift in an industrial scene are solved. According to the method, robust features are extracted through a multi-scale feature fusion auto-encoder, and a dynamic memory bank is constructed to update a normal sample prototype online; a dual-path detection mechanism is adopted to cooperate with a pixel-level reconstruction error and attention weighted feature matching deviation; efficient edge reasoning is realized in combination with block parallel processing and model compiling optimization; and designing an elastic incremental learning framework to prevent disastrous forgetting. And finally, false alarms caused by environmental changes are reduced, accurate positioning of pixel-level defects is realized, millisecond-level detection requirements of high-resolution images are met, safe and efficient model online evolution is supported, and adaptability and reliability of an industrial quality inspection system are comprehensively improved.

The invention discloses a data flow optimization method and system based on artificial intelligence. The method comprises the steps of multi-modal data flow collection, digital twin basic modeling, decision enhancement, dynamic topology optimization and resource arrangement optimization. The invention relates to the technical field of network data flow management, adopts a data flow integration optimization scheme combining digital twin basic modeling, dynamic topological optimization and decision comprehensive optimization, not only improves the flexibility and real-time performance of a system, but also optimizes the efficiency and intelligent level of resource allocation; the quantum mapping and quantum optimization technology is adopted, adaptive noise and disturbance simulation are combined, the mapping precision and the dynamic adaptive capacity between a virtual layer and a physical layer are improved, and the state error between a virtual node and a physical node is optimized through quantum mapping parallel processing; hierarchical federated learning is combined with an adaptive learning strategy to optimize a data flow decision, and a joint optimization method of a quantum heuristic algorithm and a graph neural network is introduced into dynamic topological optimization.

A new transaction barrier synchronization primitive enables executing threads and asynchronous transactions to synchronize across parallel processors. The asynchronous transactions may include transactions resulting from, for example, hardware data movement units such as direct memory units, etc. A hardware synchronization circuit may provide for the synchronization primitive to be stored in a cache memory so that barrier operations may be accelerated by the circuit. A new wait mechanism reduces software overhead associated with waiting on a barrier.