183 results about "Abstraction layer" patented technology

The invention discloses a quantum classical hybrid computing task dynamic scheduling system and method based on SLURM. The system comprises a quantum task feature extraction module, a dynamic priority evaluation module, a dependency analysis module, a quantum perception backfilling module and a uniform resource abstraction layer module. The method comprises the following steps: extracting quantum features of a to-be-processed task and a computing resource to form a quantum feature set and caching the quantum feature set; the feature set is obtained in real time, task priorities are output through multi-dimensional evaluation, and a real-time priority sequence is generated; based on the task type and the feature set, forming a dependency relationship between the classic task and the quantum task, and converting the dependency relationship into a dependency constraint and / or resource reservation instruction; according to the task priority, the resource reservation instruction and the real-time resource state, a future idle period is predicted, and a short-time quantum task is inserted for backfilling; computing resources are distributed according to the task priority and the dependency constraint, and the resource utilization state is fed back to the backfill and priority evaluation module in real time. According to the method, efficient scheduling of hybrid computing tasks can be realized, and the overall performance of the system is remarkably improved.

Technologies for creating an optimized and accelerated network pipeline using an emulated network pipeline abstraction layer (NPAL) of an emulated data processing unit (DPU), including an emulated processing device and an emulated acceleration hardware engine, are described. The emulated NPAL supports multiple network protocols and network functions in an emulated network pipeline. The emulated network pipeline includes a set of tables and logic organized in a specific order to be accelerated by the emulated acceleration hardware engine. The emulated acceleration hardware engine can process network traffic data using the emulated network pipeline.

Technologies for creating an optimized and accelerated network pipeline using a network pipeline abstraction layer (NPAL) for policy-based routing (PBR) over Service Function Chaining (SFC) are described. A DPU includes acceleration hardware engine to provide a single accelerated data plane. A processing device can generate a first virtual bridge and a second virtual bridge, the first virtual bridge to be controlled by a first network service hosted on the DPU and having a set of one or more network rules, and the second virtual bridge having a policy-based routing policy (PBR policy). The processing device can add the virtual port between the first virtual bridge and the second virtual bridge. The acceleration hardware engine, in the single accelerated data plane, can route network traffic data using the PBR policy and process the network traffic data using the set of one or more network rules.

The invention discloses a cloud server physical GPU acceleration capability release method and system, and belongs to the technical field of computer GPU capability release, and the method specifically comprises the steps: collecting GPU operation parameters in real time, carrying out the preprocessing, storing the GPU operation parameters in a shared state database, constructing a lightweight virtualization abstraction layer on a physical GPU, dividing the physical GPU into a plurality of logic units, and storing the logic units in a shared state database; the method comprises the following steps: establishing a physical GPU, constructing an isolation layer, utilizing a three-dimensional resource segmentation strategy, adopting a priority queue, a jump mechanism and a delay tolerance mechanism in a scheduling process, dynamically allocating logic unit resources of the physical GPU, accelerating release of GPU resources, and realizing unification and real-time updating of physical GPU state information among nodes of a cloud server through a distributed state synchronization mechanism; through dynamic task scheduling, a jumping mechanism, a delay tolerance mechanism and distributed state synchronization, efficient GPU resource management is realized, and GPU capability release is accelerated.

A system and method for real-time or near real-time multilingual language transformation in telecommunications environments using distributed artificial intelligence (Al). The system includes at least one processor configured to instantiate a plurality of Al agents, each corresponding to a user in a communication session, and to manage their operation via an Al-agent runtime adapter comprising a microkernel agent manager and a virtual machine layer. A language model abstraction layer enables access to language models and Al algorithm libraries through publish-subscribe interfaces. The system supports contextual transformation of speech based on user profile data such as sentiment, emotional tone, and cultural background. Execution can occur on telecommunication-class switches, edge devices, or general-purpose computing hardware. Al agents operate asynchronously, access shared or private memory, and produce personalized, bidirectional linguistic output. The architecture supports secure, scalable deployment across heterogeneous devices and networks.

Technologies for creating an optimized and accelerated network pipeline using a network pipeline abstraction layer (NPAL) are described. A DPU includes DPU hardware and memory that stores DPU software with the NPAL that supports multiple network protocols and network functions in a network pipeline. The network pipeline includes a set of tables and logic organized in a specific order to be accelerated by an acceleration hardware engine of the DPU. The acceleration hardware engine processes network traffic data using the network pipeline.

The present application provides a large language model-based query statement generation method, apparatus, and device, and a medium. The large language model comprises a plurality of encoders. The method comprises: acquiring a query text; performing vectorization processing on the query text to obtain a target vector representation corresponding to the query text; inputting the target vector representation into the plurality of encoders, to perform encoding processing on the target vector representation by means of weight matrices of the encoders, so as to obtain context vectors corresponding to the encoders, wherein different encoders have different weight matrices; and generating a target query statement on the basis of the context vectors corresponding to the encoders. A model is allowed to focus on different information features in different attention heads, independently capture different aspects of a query text, and process information at multiple abstraction levels, thereby better capturing complex and abstract semantic relationships, effectively distinguishing the importance of information when processing complex and fuzzy questions, improving the accuracy of query statement generation.

The invention relates to code verification, in particular to a code and document consistency reasoning verification system based on a cross-modal logic atlas, which comprises an input layer for receiving source codes and software documents; the feature extraction layer is used for performing feature extraction on the source code and the software document; the abstraction layer is used for converting the extracted features into comparable intermediate representations; the unified modeling layer is used for mapping code logic and document semantics into a unified atlas structure and performing formalized representation by using description logic to obtain a cross-modal logic atlas; the reasoning verification layer is used for performing reasoning verification on the consistency of the code and the document according to the cross-modal logic graph; the result processing layer is used for judging a verification result according to the reasoning verification condition, starting a contradiction positioning mechanism when inconsistency is found, and carrying out contradiction positioning and report generation to obtain an interpretable report; according to the technical scheme provided by the invention, the defects that the consistency between the code and the document is difficult to accurately verify, and accurate contradiction positioning and an interpretable report are difficult to provide can be effectively overcome.

The invention relates to the technical field of business process management, in particular to a data processing method, system and device based on business process management and a medium, and the method comprises the steps: constructing a business process model driven by an abstraction layer, and configuring a multi-protocol standardized interface; according to the business process model driven by the abstraction layer and the multi-protocol standardized interface, a process instance matched with the business scene is generated, and a task is distributed to a target client according to a task demand; managing form state circulation and node jump logic according to a task control engine; an event monitor is deployed through a process key node to monitor the process, and when the process reaches a specified node, external service calling operation is triggered to form an end-to-end automatic closed loop. Therefore, the problems of code redundancy, high maintenance cost, difficulty in quickly adapting to diversified customer requirements and the like due to the fact that business process management generally needs to be subjected to customized development for different subsystems in related technologies are solved.

The invention relates to the technical field of deep learning, in particular to a PyTorch framework-based compatibility expansion system, which comprises a cross-framework model converter, a heterogeneous hardware abstraction layer, a hybrid computational graph execution engine, an intelligent distributed trainer and a self-adaptive optimizer, the cross-frame model converter converts input model formats of different frames into PyTorch executable formats, the heterogeneous hardware abstraction layer supports rear ends of various hardware and automatically selects an optimal calculation path, and the hybrid calculation graph execution engine fuses dynamic graph flexibility and static subgraph optimization and supports dynamic control function execution flow. An intelligent distributed trainer automatically selects a parallel scheme and is compatible with multi-protocol communication, a self-adaptive optimizer performs dynamic optimization based on hardware characteristics and a PyTorch tensor, a cross-frame model converter supports model conversion of multiple deep learning frames, the cost of migration among different frames by a user is reduced, and the universality of the PyTorch frame is improved.

The invention discloses an intelligent routing and unified adaptation method for a large language model, and the method comprises the steps: defining all access details of the model through a declarative configuration file, and achieving the zero-code access of the large language model without writing any code for a newly-added model; a completely consistent calling interface is provided for an upstream application, the isomerism of all downstream large language models is shielded, and a unified request and response abstraction layer is constructed; through a strategy engine and a JSON path technology, complex streaming response including content thinking is precisely processed, and intelligent analysis and content extraction are carried out; dynamic configuration and intelligent strategy hot update are supported; intelligent routing of the model is realized, and an optimal large language model instance is dynamically selected; and meanwhile, enterprise-level governance capability is provided, governance functions such as fusing, degradation, current limiting and monitoring are integrated, and stability guarantee is provided for model calling. Therefore, the maintainability, the expandability and the user experience consistency of the system are comprehensively improved.

A cloud computing platform provides zero trust network access as a service to customers that maintain applications on-premises. In this context, the cloud computing platform may associate customers and / or applications with specific service proxies, and add an abstraction layer for network access that maps an alias domain for each customer and / or application to a network load balancer associated with the specific service proxies associated with the corresponding application(s). This approach advantageously simplifies the configuration of service proxies at the cloud computing platform by permitting dedicated relationships among network load balancers, specific service proxies, and specific applications, while concurrently reducing or avoiding the administrative burden on customers of updating network pointers when the clusters of service proxies are periodically reconfigured to adjust to varying user traffic.

A system and method of decoding a bitstream to produce tensors for use by a neural network second portion. The method comprises decoding a network abstraction layer (NAL) unit from the bitstream having a predetermined length, wherein the NAL unit of the predetermined length indicates a network abstraction layer (NAL) unit format of one inner codec of a plurality of inner codecs, each other inner codec having NAL unit lengths different to the predetermined length; selecting an inner codec from the plurality of inner codecs based on the decoded NAL unit of the predetermined length; and decoding the bitstream using the selected inner codec to produce the tensors.

The invention relates to the technical field of Internet of Things, in particular to an Internet of Things equipment collaborative management system based on an intelligent edge, which is composed of an equipment abstraction layer, a converged communication layer, a service logic layer and a firmware upgrading layer, the equipment abstraction layer realizes standardized access and full life cycle management of equipment through modeling, interface abstraction and resource pool management; the converged communication layer comprises protocol adaptation, intelligent routing and QoS guarantee to ensure smooth multi-protocol communication; the business logic layer adopts a micro-service architecture, and flexible business processing is realized through flow arrangement, intelligent decision making and event driving; the firmware upgrading layer comprises state perception, intelligent analysis, intelligent execution and firmware management so as to adaptively select an optimal upgrading strategy according to network state prediction and evaluation. According to the invention, modules at all levels cooperate, intelligent and collaborative management of Internet of Things equipment is realized, and the equipment management efficiency and the operation stability are improved.

The invention discloses a construction method for dynamically reconstructing a multi-modal edge computing hardware architecture and a cross-industry application method, and relates to the field of edge computing hardware architectures. A five-level heterogeneous architecture is adopted to construct a field self-adaptive dynamic reconstruction calculation unit, and an equipment abstraction layer is constructed to realize module hot plugging and calculation power distribution; a multi-protocol ubiquitous interconnection interface engine is established, a hardware-level analysis architecture is integrated, and multi-protocol analysis and rapid switching are supported; an environment self-adaptive energy efficiency regulation and control system is deployed, and wide-temperature operation is realized by adopting a bionic composite heat dissipation design; an edge-region-cloud three-layer elastic scheduling network is constructed, cross-domain computing power arrangement is realized through federal learning, and the method is adaptive to multi-industry scenes. According to the method, the hardware reuse rate and the wide-temperature environment adaptability are improved, the multi-protocol processing delay is reduced, the computing power scheduling efficiency is optimized, the limitation of traditional equipment is broken through, and the efficient and stable operation requirements of cross-industry scenes such as electric power and medical treatment are met.

The invention relates to the technical field of computer server resource management and network virtualization, and particularly discloses a server hybrid deployment and management system based on a virtualization technology, which comprises a uniform resource abstraction layer, a global resource sensing and modeling unit, a strategy-driven intelligent arrangement engine and a heterogeneous resource execution adapter. The uniform resource abstraction layer performs standardized abstraction on a physical server, a virtual machine and a container; the global resource sensing and modeling unit integrates the resource data and constructs a global resource model; a strategy-driven intelligent arrangement engine performs multi-objective optimization solution based on strategies and constraints to generate a scheduling scheme; the heterogeneous resource execution adapter translates the scheme into bottom executable atomic operations. According to the invention, unified management and intelligent cooperative scheduling of heterogeneous computing resources are realized, and the resource utilization efficiency and the system automation level are improved.

The invention discloses a data layout optimization method and device applied to NPU code compilation and a medium. The method comprises the steps that an intermediate representation IR input by an upper layer is split into an operation type OP and an operand, and the operand input by the upper layer is divided into logic data and a logic mask; converting the logic data of the upper layer into data vector representation of the abstraction layer, and converting the logic mask of the upper layer into mask vector representation of the abstraction layer; obtaining a corresponding bottom hardware instruction capability according to an operation type OP input by an upper layer, and performing legalization operation on the operation type OP; and performing legalization processing on the converted abstract data vector type representation and mask vector representation according to the target hardware capability, performing instruction mapping on the processed abstract instruction according to the underlying hardware pair operation type OP, and generating a target LLVM instruction compatible with the underlying hardware. Register layout abstraction and conversion can be carried out on the NPU code under mixed precision input, and the utilization efficiency of the NPU bottom layer register is improved.

Device for the intelligent encapsulation of legacy data information systems with the integration of artificial intelligence, comprising: a collection unit for legacy interfaces, configured to intercept communication data from existing platforms; a semantic interpretation unit designed to analyze and map data structures and process semantics; a contextual abstraction layer designed to convert interpreted information into standardized, interoperable formats; a policy-driven management module that ensures secure access, Data transformation and regulatory compliance are ensured; and a learning and inference unit for adaptive optimization of data mappings and operational logic.

The invention discloses a natural resource asset checking management system and method based on a cross-platform architecture, and relates to the technical field of resource management, and the system comprises a cross-platform adaptive architecture module, a real-time data synchronization and verification module and an intelligent analysis and multi-dimensional visualization module. The cross-platform adaptive architecture module carries out fragmentation preprocessing on collected data through a uniform interface abstraction layer, a dynamic adaptation layer, a cross-platform rendering engine and a platform difference compensation mechanism and the real-time data synchronization and verification module, and the transmission efficiency is optimized through multi-link transmission and dynamic path selection. And the timeliness and accuracy of data synchronization are ensured in combination with block chain hash check and conflict detection. According to the system, through cross-platform adaptation, efficient data synchronization and intelligent visual analysis, the efficiency and scientificity of natural resource asset checking management are remarkably improved.

The invention discloses a method and system for packaging a webpage embedded sign language translation JS library, and belongs to the field of electrical digital data processing.The method comprises the steps that by constructing an ECS core architecture with perception feedback, a bidirectional data flow between modules is established to achieve dynamic collaboration of rendering and grammar; predictive streaming resource management is adopted, and resources are intelligently preloaded based on gesture sequence contexts; multi-instance physical isolation is realized by utilizing a sandboxed container and a ShadowDOM (Document Object Model); implementing multi-level adaptive rendering, and dynamically switching a rendering path and a frame rate according to equipment performance; and finally, shielding the bottom layer difference through cross-end unified rendering of the abstraction layer. According to the method, the problems of animation and semantic dislocation, resource loading conflict and poor equipment compatibility in a multi-user concurrent scene are effectively solved, and high-fidelity, low-delay and cross-end consistent experience of sign language translation animation is realized.

The invention relates to the technical field of automatic data processing of integrated circuit design, in particular to an automatic parasitic parameter result analysis system based on an FUZZ test. The fuzzy test engine module adopts syntax tree variation to generate a test case library comprising illegal characters, extreme values and logic loops, the exception response module captures illegal operations in an analysis and simulation stage through the probe detection unit, and the correction decision unit executes data restoration or generates an exception report. And the incremental updating unit synchronizes the abnormal features to a model self-optimization module rule base. And the dynamic adaptation module converts the multi-process parasitic parameters into a standardized simulation model through the process parameter abstraction layer. All the modules form a closed-loop system for test case generation, anomaly detection and repair and rule dynamic optimization, automatic processing of an unexpected input scene is achieved, and robustness and multi-process compatibility of parasitic parameter analysis are improved.

The invention provides a server delivery and operation and maintenance automation system oriented to multi-platform collaboration, and aims to solve the problems of split delivery process, lagging expansion response, high operation and maintenance complexity and the like in a heterogeneous infrastructure environment. According to the system, by constructing a multi-platform adaptive abstraction layer, bottom layer differences of private cloud, job scheduling platforms and the like are shielded, and unified modeling and standardization of delivery and operation and maintenance processes are achieved; on the basis, key links such as installation file generation, expansion file generation, configuration injection and deployment verification are integrated, an end-to-end one-key delivery assembly line is built, and rapid and consistent deployment of a cross-platform environment is supported; meanwhile, an automatic operation and maintenance closed loop is constructed, and operation and maintenance operations such as health examination, fault warning, configuration repair, service restart, node replacement and elastic capacity expansion and contraction are automatically executed by collecting operation states, logs and performance indexes of the server and the service in real time and combining a preset strategy or an intelligent analysis model. By means of the method, efficient delivery and intelligent capacity expansion of server resources are achieved, and the autonomous operation and maintenance capacity, the high availability and the service continuity guarantee level of the system in a complex multi-platform mixed deployment scene are remarkably improved.

The invention discloses a test case editing scheme oriented to the automatic test industry, and the scheme realizes efficient editing, analysis and execution of test cases through layered architecture design. The scheme comprises the following steps: (1) dividing a test case into a measurement abstraction layer and a hardware abstraction layer; (2) in the measurement abstraction layer, constructing a test index sequence and a corresponding index operation sequence to form a test process file; (3) in the hardware abstraction layer, establishing a mapping relation between the operation keyword and the underlying instrument driver, and forming an instrument configuration file; and (4) the test process file and the instrument configuration file are converted into objectified description, so that flexible editing and automatic script conversion of the test case are realized. The method has the beneficial effects that (1) the readability and editing efficiency of the test case are improved, and the workload of manual script conversion is reduced; (2) the generalization ability of the test scheme is enhanced, and the method is suitable for various test devices and test scenes; and (3) flexible configuration and efficient reuse of test resources are realized through a layered architecture and a standardized file format.

The invention discloses a domain controller automatic test scheduling method and system based on a hierarchical software architecture, and relates to the technical field of automatic test and industrial control, and the method comprises the steps: S1, constructing an automatic test system of the hierarchical software architecture which at least comprises a hardware drive abstraction layer and a test application layer, the hardware drive abstraction layer is used for uniformly packaging bottom layer drives of a board card, a signal conditioning module, a relay matrix and a power supply module in the test system, and providing a standardized interface irrelevant to specific hardware for the test application layer; according to the domain controller automatic test scheduling method and system based on the hierarchical software architecture, the comprehensive requirements of a new energy automobile power domain controller on high-efficiency, high-reliability and high-consistency automatic test in research, development, verification and mass production stages are better met.

The invention discloses an AI collaborative interaction system and device of a cross-platform standardized database, and relates to the technical field of AI collaborative interaction systems, the AI collaborative interaction system comprises a data processing module, a standardization module, an interface module, a data management module, a prediction module and an interaction feedback module; the data processing module is used for collecting original data by adopting a distributed crawler technology to obtain a basic data set; the standardization module is used for carrying out standardization processing on the basic data set to obtain a standardized data packet; the interface module is used for performing cross-platform database access operation on the standardized data packet by adopting an abstraction layer design method to obtain a cross-platform query result; the cross-platform query result obtained by the interface module is analyzed and optimized by adopting the machine learning algorithm, the query performance is improved, the system can automatically identify and optimize the query path through the application of the machine learning algorithm, and the query efficiency is remarkably improved.

The invention discloses an RoCEv2 protocol RC service type basic primitive function verification system based on a UVM. The system comprises a test case class used for configuring verification parameters, initializing a verification environment and driving a primitive execution process; the verification environment is used for packaging a plurality of agent components, a reference model, a register model and a scoreboard and carrying out data interaction through a TLM interface; the agent components are used for driving or monitoring an interface designed to be tested, comprise but are not limited to a register read-write agent, a work queue element WQE reading agent, a load data reading and receiving agent, a data packet sending and receiving agent, a complete queue element CQE receiving agent and a register model component, and are constructed based on a register abstraction layer RAL modeling method; and front-door and back-door access paths are supported. The method is high in automation degree, integrates constraint randomization, comparison verification and coverage rate analysis, and realizes full automation of the verification process.

The invention relates to the technical field of computer authority management, and discloses an office authority transfer management system for mobile office, which comprises an authority abstraction layer module used for converting original authority representation of each terminal platform into unified and standardized intermediate state representation; the permission authorization service module is used for supporting permission verification, granting and revocation operations in an offline state; the scene sensing module is used for constructing a multi-dimensional scene feature space and identifying key features related to safety in a current scene through a multi-head attention network; the permission adjusting module is used for realizing a continuous permission control strategy from strict limitation to high freedom; the behavior monitoring module is used for recording permission use behaviors in an offline state and detecting potential abnormal operations; according to the method, the working continuity and the user experience in a mobile office environment are improved, and meanwhile, the enterprise data security is effectively guaranteed through a permission progressive release mechanism and a permission re-evaluation mechanism triggered by environment change.

The invention belongs to the technical field of financial information security, and discloses a financial data leakproof terminal protection system based on behavior biological recognition, which comprises a cross-platform behavior acquisition module, and is characterized in that compatible adaptation of a PC terminal, a mobile terminal and a remote desktop terminal is realized through a unified behavior acquisition abstraction layer; the method comprises the following steps: acquiring user operation behavior characteristics in real time by adopting a cross-platform container technology and a native plug-in interface, and performing pre-encryption processing and structured coding to form a user behavior sequence; the cognitive trajectory recognition module is used for constructing individual behavior cognitive vectors based on a historical user behavior sequence and generating a user cognitive signature path by adopting self-supervised learning; generating a preliminary risk tag by comparing the current user behavior sequence with the cognitive signature path; identifying a user abnormal behavior track, and constructing a user abnormal track graph structure by extracting user abnormal behavior track features; the security protection level of the terminal is effectively improved, and the terminal adapts to complex and changeable security threat environments.

The invention relates to the technical field of digital twin and multi-source heterogeneous data fusion, and discloses a method for loading and rendering massive homologous heterogeneous data of a digital twin platform. The method aims at solving the problems of low loading efficiency, large rendering fluctuation, data semantic segmentation and the like in the prior art. The technical scheme comprises the steps of preprocessing data and establishing a correlation index; during initialization, a network state is adapted to load basic data; constructing unified abstraction layer standardized layer management; on-demand loading and multi-level caching are realized through the intelligent scheduler; a fusion rendering pipeline is configured to guarantee stable output of the 60FPS; operating a service binding engine to realize bidirectional linkage; measurement, model import, file preview, layer linkage and data maintenance are supported. The mass data loading efficiency is improved, memory overflow is avoided, service collaboration and system stability are enhanced, and the method is suitable for scenes such as urban planning, industrial simulation and emergency command.

The present invention provides a device, a computer-readable storage medium, and a method for exchanging data between containers, all equipped with shared memory. [Solution] Embodiments of this specification describe a Platform Abstraction Layer (PAL) for edge devices that enables services developed in different programming languages ​​to run as a pipeline on edge devices (e.g., cameras). For example, services may be deployed in separate containers developed using different programming languages. In one embodiment, stages in the pipeline can use buffers in shared memory to share data. The data may include a unique ID (e.g., frame ID) to identify the buffer containing the processed data by searching an index array for the shared memory. In this way, the shared memory supports random access.