346 results about "PERQ" patented technology

The invention discloses a data center machine room AI energy-saving control method and system, a digital twin model of a machine room operation state is constructed through a holographic perception and heterogeneous data fusion technology, centimeter-level monitoring of an equipment state and environmental parameters is realized, and the system integrates a laser radar array, an acoustic sensor and a gas sensor network. The time-space alignment of multi-modal data is completed by combining edge computing nodes, holographic mapping including thermodynamic characteristics, vibration characteristics and gas leakage risks is formed, historical temperature control strategy characteristics are extracted by adopting a variational auto-encoder based on a dynamic strategy generation mechanism of generative artificial intelligence, and a load trend is predicted by combining a long-short-term memory network. Constructing a self-adaptive strategy pool; the multi-agent reinforcement learning framework enables temperature control, equipment scheduling and power grid response to form game optimization, the strategy robustness in a complex scene is improved, and the system innovatively fuses power grid real-time electricity price and carbon transaction data so as to establish a multi-target decision system.

The invention provides a Linux access control system based on attributes, and relates to the technical field of data access control. The system comprises a system monitor module, a data interaction module and a decision unit. The system monitor module collects attributes from a kernel and a user space and writes the attributes into the data interaction module; the access decision unit compiles the access control strategy into an eBPF program and mounts the eBPF program to a corresponding hook; executing the kernel to the hook, and triggering an eBPF program; an eBPF program queries a Flow rule; matching the attribute with the Flow rule, and if the matching is successful, executing a corresponding action; if all the Flow rules fail to match, executing a default action; the system can be expanded during operation, and can be loaded or unloaded based on dynamic loading characteristics and strategies of the eBPF program and the eBPF program during operation of the system, so that the problem that a kernel needs to be compiled in a traditional LSM scheme is solved; the method does not intrude the kernel, is completely based on an eBPF program, does not modify a kernel source code, and can guarantee the stability and compatibility.

The invention relates to the technical field of virtualization management, in particular to a virtual machine operation management system based on RISC-V. The system comprises an instruction set mapping module, a resource scheduling optimization module, a virtual machine monitoring integration module, an operation tuning update module and an operation configuration distribution module. According to the method, by analyzing the operation instruction of the virtual machine and combining the RISC-V underlying characteristics, accurate instruction mapping is achieved, the processing efficiency and accuracy are remarkably improved, the performance loss and conversion delay caused by too high abstraction level of a traditional virtual machine on the instruction level are overcome, the optimal allocation scheme is screened, the resource utilization rate and the system response speed are effectively improved, and the system performance is improved. The operation state and performance indexes of the virtual machine are integrated, anomaly detection and resource recovery are embedded, system stability and effective resource reutilization are ensured, basic configuration is continuously and incrementally updated, operation configuration refinement and automatic management and updating are achieved, the overall operation efficiency and management flexibility are improved, and the manual intervention cost is reduced.

The invention discloses a power consumption and performance balanced scheduling method and system for a heterogeneous multi-core processor, and belongs to the technical field of computers, and the method comprises the steps: obtaining historical load data and current system state parameters, generating a load sequence and carrying out load prediction analysis, generating a predicted load value for feature fusion and quantification, and generating a task feature vector; constructing a core capability portrait, calculating a matching degree between the task characteristic vector and the core capability portrait, and generating a scheduling strategy; and dynamically combining the physical cores into a logic calculation unit according to a scheduling strategy, configuring a power consumption state, and executing task allocation and migration. According to the method, the technical means of load prediction analysis, quantitative matching of tasks and cores and dynamic combination and configuration of physical cores are adopted, and pre-judgment and prospective distribution of system resources can be achieved, so that performance requirements are met, meanwhile, the system energy efficiency is effectively optimized, and deep balance of power consumption and performance of a processor is achieved.

The invention discloses a deployment method and device of a large model agent, equipment and a medium, and relates to the technical field of artificial intelligence, and the method comprises the steps: determining agent configuration information of a to-be-deployed large model agent after an agent deployment request is received, and determining a target large model and model metadata; determining an available resource condition of a preset computing resource pool, and performing query matching according to the available resource condition and the model metadata to determine a plurality of hardware-compatible candidate computing resource nodes; performing load prediction operation on the to-be-deployed large model agent after the large model agent is online to obtain a predicted load, and generating a target deployment decision scheme based on the predicted load, a preset scheduling strategy, the available resource condition and the candidate computing resource node; and adjusting the runtime parameters of the target large model according to the resource characteristics of the target computing resource node, and loading the adjusted target large model on the target computing resource node so as to deploy and start the agent instance of the to-be-deployed large model agent.

The invention provides a method, a system and equipment for improving the computing performance stability of a hardware platform and a medium, and belongs to the technical field of computer system performance optimization. The method comprises the following steps: acquiring architecture characteristic data of a bottom hardware platform through a hardware detection tool, and generating and loading a first-stage system kernel parameter configuration based on the architecture characteristic data; reading the architecture characteristic data, carrying out behavior analysis and type identification on the running process based on the architecture characteristic data, and generating a second-level process scheduling strategy according to the behavior analysis and type identification; reading cache structure information in the architecture characteristic data, and driving a file system to intelligently prefetch and reconstruct a storage layout of memory data based on the cache structure information; and continuously collecting performance index flow data when the system runs, performing real-time analysis on the performance index flow data by using a performance degradation model, generating a feedback control instruction according to an analysis result, and dynamically correcting the kernel parameter configuration of the first-stage system and the scheduling strategy of the second-stage process.

The invention discloses a heterogeneous resource energy-saving optimization scheduling method and system based on computing power perception, and relates to the technical field of modern computing infrastructure, the system comprises a computing power-energy consumption evaluation module, the computing power-energy consumption evaluation module is used for carrying out computing power and energy consumption evaluation on heterogeneous resources, generating and dynamically updating a computing power-energy consumption mapping table, the heterogeneous resources comprise ARM architecture resources, X86 architecture resources and GPU resources. According to the method, the computing power-energy consumption mapping table is generated and updated in real time, so that the scheduling decision is always based on the latest energy efficiency state of the resources, the accuracy of the basic data is ensured, intelligent task classification and dynamic constraint construction are combined, the system can flexibly balance the task performance and the energy consumption target, and particularly, the strategy can be adaptively adjusted for a high-emergency task; the timeliness of key services is guaranteed, the differentiated scheduling strategy serves as an efficient pre-screening mechanism, task requirements and resource architecture characteristics are accurately matched, and the pertinence and efficiency of resource allocation are remarkably improved.

The invention discloses a controller programming method and device based on an equipment model, equipment, a medium and a product, the equipment model is used as a unified semantic anchor point, heterogeneous characteristics of physical equipment are converted into computable structured description, hardware adaptation and business logic are decoupled through a driving framework, a large model focuses on generation of a high-value control strategy, and the control strategy is optimized. Semantic drift and logic illusion of a program generation large model are remarkably inhibited by relying on double constraints of equipment document data and a knowledge graph, finally, an executable drive program meeting strict requirements of an industrial site is output through compiling verification of a closed loop, and due to the fact that manual development is not needed, development efficiency is improved, development cost is reduced, and development efficiency is improved. The equipment model provides a cross-protocol and cross-manufacturer standardized interface contract, so that the problem of poor reusability in traditional development is solved, and the reusability is improved.

The invention provides a cloud edge collaboration algorithm arrangement and deployment method, system and device and a storage medium, and relates to the technical field of cloud edge collaboration.The method comprises the steps that algorithm task requirements input by a user are analyzed, and after corresponding task information is obtained, an intermediate representation file is generated; collecting computing power state information of the plurality of edge devices to determine a target edge device used for executing an algorithm link, and generating an operation parameter configuration file according to a hardware architecture of the target edge device and algorithm characteristics of the intermediate representation file; and issuing the intermediate representation file and the operation parameter configuration file to the target edge device, and performing automatic compiling and deployment according to the chip type of the target edge device. By issuing the intermediate representation file and the operation parameter configuration file to the target edge device and automatically executing compiling and deployment, the effects of cross-platform automatic adaptation, rapid deployment and efficient operation are achieved, and the problem that in the prior art, task organization and deployment flexibility is insufficient is solved.

The invention relates to the technical field of engineering simulation, and discloses a GPU heterogeneous resource scheduling method based on mixed execution of multiple simulation jobs, which considers the calculation characteristics of different simulation tasks, also introduces a calculation time prediction model and a video memory usage amount prediction model, and prejudges the resource consumption condition during job execution in advance. And accurate depiction of resource demands in a complex multi-physics field simulation scene is realized. When potential conflicts are found, a Monte Carlo tree search technology is utilized to explore a plurality of scheduling strategies, and a scheduling scheme which achieves optimal balance among calculation delay, throughput and energy consumption is screened out from the scheduling strategies. According to the method, historical data and real-time job requirements of a simulation platform are comprehensively analyzed, accurate prediction and dynamic allocation of GPU heterogeneous resources are achieved, the utilization rate and the overall performance of the GPU heterogeneous resources in a multi-simulation job mixed execution environment are effectively improved, and support can be provided for efficient resource scheduling in a complex simulation scene.

The invention discloses a hierarchical self-adaptive full block pre-filling scheduling method and system for large language model reasoning, and the method comprises the steps: carrying out the hierarchical portrait analysis of a to-be-served model, and dividing the to-be-served model into partitions with different calculation characteristics according to the calculation intensity and memory access characteristics of each layer; then, making a layering and partitioning strategy based on a partitioning result, allocating a larger partitioning size to a calculation-intensive partition, allocating a smaller partitioning size to a memory bandwidth-intensive partition, and generating a layering and partitioning mapping table; and finally, when the online scheduling is executed, querying the mapping table according to the request processing progress to determine the block target size, and jointly forming a batch processing unit by the decoding task and the pre-filled block with the heterogeneous size under the constraint of the iteration time budget to be executed. According to the method, accurate matching of calculation and bandwidth resources is achieved, the system throughput can be effectively improved, tail delay and fluctuation thereof can be remarkably reduced, bubbles under pipeline parallelism are reduced, and the method is suitable for various attention mechanisms and distributed reasoning architectures.

The invention discloses a digital twinning three-dimensional model dynamic synthesis system based on multi-modal feature fusion, and belongs to the technical field of digital twinning, computer processing and three-dimensional reconstruction crossing. Data flow characteristics are divided into high-frequency real-time data flow in a machine tool running state and low-frequency / event triggering data in a machine tool static state, a specific processing fusion path is intelligently selected based on a self-adaptive double-path processing mechanism, a state parameter set for driving a three-dimensional basic model is generated, and the three-dimensional basic model is obtained. The method not only optimizes the dynamic synchronization and prediction of the operation state, but also achieves the modeling and tracking of the static state (such as cooling recovery, maintenance calibration and environmental evolution) of the machine tool, constructs a digital twin system covering the whole scene of operation-shutdown-maintenance of the numerical control machine tool, and supports the real whole life cycle management.

The invention discloses a satellite communication terminal application software modular design method, which is suitable for an ARM (Advanced RISC Machines) architecture Linux platform and aims to solve the core defects of traditional satellite terminal software. The method comprises the following steps of: firstly, dividing 14 independent modules by combining platform characteristics and business requirements, and interacting the modules through a standard interface adaptive to the bottom specification of the platform; secondly, a main process and multiple sub-threads framework is adopted, the main process initializes system resources and pulls up the sub-threads of all modules, and the sub-threads support dynamic loading / unloading and fault restarting through IPC cooperation, so that overall collapse is avoided; and finally, abstracting driving logic through a hardware interface module, realizing multi-hardware compatibility in combination with a common function of a common module, and monitoring the state by means of an alarm and log module to ensure stable operation.

The invention provides a heterogeneous multi-agent autonomous collaborative optimization method for production and maintenance integrated scheduling of a mixed flow manufacturing workshop. The heterogeneous multi-agent autonomous collaborative optimization method comprises the steps of limited buffer region constraint, feature screening, heterogeneous multi-agent and the like. The method is optimized on the basis of a TD3 algorithm, and comprises the following steps: firstly, constructing a heterogeneous multi-agent according to the characteristics of each processing site; secondly, in order to improve effective characterization of feature extraction, a feature screening mechanism based on NSGAIII is designed; and finally, in combination with the actual scheduling condition of the mixed-flow manufacturing workshop and aiming at the actual situation that the performance of the scheduling rule of the first-selected workpiece and the second-selected machine is poor, a composite scheduling rule based on the first-selected machine and the second-selected workpiece is adopted. According to the method provided by the invention, the effectiveness and superiority of the NSGAII-MATD3 algorithm are respectively demonstrated by comparing a traditional single scheduling rule, a feature-free screening MATD3 algorithm and five advanced deep reinforcement learning algorithms.

The invention provides a test program generation method, electronic equipment and a storage medium, relates to the field of chips, and is mainly technically characterized by comprising the following steps: generating a plurality of subprograms corresponding to each module in at least two modules according to instruction characteristics of the at least two modules in a target processor; wherein each subprogram comprises dependency information with other subprograms, each subprogram comprises a plurality of instruction fragments, each instruction fragment comprises at least one instruction, and the dependency information of the subprogram is used for indicating whether the instruction fragment in the subprogram conflicts with the instruction fragments in other subprograms or not; the to-be-used instruction fragments are continuously selected until a preset ending condition is met, and the next to-be-used instruction fragment in every two adjacent to-be-used instruction fragments is the to-be-used instruction fragment which is sequentially selected from any subprogram of any module and does not conflict with the previous to-be-used instruction fragment; and obtaining a target test program according to a selection result of the to-be-used instruction fragment.

The invention provides an intelligent data management method and system, and relates to the technical field of data processing. Standardized packaging and automatic execution of the data governance instruction are achieved, and the problem that a traditional storage medium cannot dynamically adapt to the multi-source data processing requirement is solved. Through the decoupling design of the pre-compilation instruction and the runtime environment, the same storage medium can adapt to data management nodes of different hardware architectures, and the deployment complexity during system upgrading is effectively reduced. Furthermore, based on an instruction execution mechanism of a streaming computing engine, the real-time performance of data quality evaluation and early warning triggering is ensured, and decision lag caused by batch processing delay is avoided. Through the on-demand loading characteristic of the modular instruction set, the function expansion requirement of a specific business scene is supported while the core governance function is ensured.

The invention relates to an RRC (Radio Resource Control) layer enhanced mobile model simulation method and system based on a 3GPP (3rd Generation Partnership Project) NTN (Network Temporary Network) standard. The method comprises the following steps: S1, constructing a simulation environment conforming to the 3GPP NTN standard; s2, constructing an RRC state machine in the NTN scene, wherein the RRC state machine comprises a user RRC state machine and a satellite RRC state machine; s3, configuring a dynamic conversion condition of the RRC state machine and a corresponding behavior logic; s4, scheduling events and injecting NTN parameters, and performing NTN propagation characteristic simulation based on the simulation environment; and S5, counting performance indexes generated in the simulation process, and outputting an NTN scene simulation report. According to the method, the simulation environment conforming to the 3GPP NTN standard is constructed, the RRC state machine in the NTN scene is constructed, the characteristics of long delay, high mobility and orbit prediction of the satellite are adapted, and the problem that a state machine model of a traditional simulation method is not applicable is solved.

The invention relates to the technical field of power equipment inspection data processing methods, in particular to a Grouping-DINO model acceleration method combined with a multi-stage progressive multi-mode distillation mechanism. According to the method, through the whole process of data set construction, lightweight model design, three-stage progressive multi-mode distillation and deployment optimization, the core pain point of electric power inspection is systematically solved. Firstly, a training foundation is built by using an electric power scene adaptive data set, and then a lightweight student model conforming to edge equipment computing power limitation is designed in a targeted manner, so that the problem of computing power mismatching of a general model is avoided; key points are that teacher model capability is migrated by stages by means of a three-stage distillation mechanism, lightweight precision loss is made up in combination with special optimization for an electric power scene, and edge equipment is further adapted through deployment optimization; finally, the technical problem that the routing inspection detection precision is reduced due to the fact that adaptation optimization is not carried out for the layered structure, the shielding characteristic and the terminology instruction of the power equipment is solved.

An embodiment may involve receiving a textual description of a workflow; determining, based on the textual description and operational characteristics associated with a computing system, a workflow outline that represents respective steps of the workflow; for each of the respective steps of the workflow, identifying a respective portion of the operational characteristics based on determining that the respective portion satisfies a relevance criterion; obtaining, from a natural language model and based on the respective portions of the operational characteristics, a respective programmatic representation for each of the respective steps of the workflow; and updating the workflow outline based on the respective programmatic representations.

The invention discloses a GPU cluster resource allocation method and device, equipment and a storage medium, and relates to the technical field of GPU cluster resource allocation. According to the method, a hardware performance description vector is generated by fusing static hardware parameters and dynamic micro-benchmark test data, the limitation that the GPU performance is represented only by depending on the static parameters is broken through, and the actual operation capacity of different architecture GPUs in a heterogeneous cluster is accurately matched; task feature vectors are generated by extracting task calculation features and resource demands, and quantitative description of reasoning task resource consumption features is achieved; information of hardware, tasks and load dimensions is integrated through a machine learning model, performance degradation characteristics during multi-task parallel are effectively captured, and the accuracy of execution time prediction is improved; scheduling decision operation including candidate node screening, comprehensive cost evaluation and resource reservation backfilling is executed in combination with the predicted execution time, and the task execution efficiency and cluster load balancing are both considered.

The invention provides an OpenVX framework system for NPU acceleration, which comprises an application layer, an OpenVX framework layer, an NPU runtime layer and an NPU hardware layer, and is characterized in that the OpenVX framework layer comprises an NPU perception graph optimizer. According to the invention, a set of OpenVX framework for completing the visual task on the NPU acceleration chip is designed, the calculation graph of the visual task constructed by the standard OpenVX API can be optimized according to the hardware characteristics of the NPU, and the memory management of the hardware layer of the NPU is designed, so that the visual task can be completed on the NPU in an accelerated manner, and the calculation efficiency of the visual task is improved.

The invention relates to the technical field of computer hardware resource management, in particular to a dynamic load balancing optimization system for a computer mainboard, which comprises a data acquisition module, a feature decoupling module, a load evaluation module, a topology construction module and a strategy generation module. The system collects real-time load data of hardware components of a mainboard, and calculates intensive, data throughput and input and output blocking load features through feature decoupling separation. And generating a load pressure index and a migration potential coefficient of each component based on the decoupling characteristics, and dynamically constructing a mainboard resource topological graph. The system simulates the influence of different load migration schemes on delay and energy consumption on a topological graph, and generates a collaborative optimization instruction set integrating task scheduling, frequency adjustment and power distribution strategies. According to the method, fine identification of load characteristics and global collaborative scheduling of system resources are realized, and the resource utilization efficiency and the energy efficiency ratio are improved.

An embodiment may involve receiving a textual description of a workflow; determining, based on the textual description and. operational characteristics associated with a computing system, a workflow outline that represents respective steps of the workflow; for each of the respective steps of the workflow, identifying a respective portion of the operational characteristics based on determining that the respective portion satisfies a relevance criterion; obtaining, from a natural language model and based on the respective portions of the operational characteristics, a respective programmatic representation for each of the respective steps of the workflow'; and updating the workflow outline based on the respective programmatic representations.

The invention belongs to the field of end-cloud collaborative computing and lightweight virtualization, and provides an end-cloud collaborative lightweight application distribution and isolated operation method and system, which adopts a WASM byte code as a unique format for application distribution, compiles an application into a WASM module at a cloud and then distributes the WASM module to an end side device; the end side equipment combines read-only configuration data of a host with a temporary file system Tmpfs of a memory by utilizing OverlayFS on a file level, and maps the data into a virtual file system in a WASM sandbox; on the symbol level, a dynamic WASI bridge is introduced, according to the requirement of an import section in a WASM module, the native capacity of a host system is injected into WASM operation in a host function mode in real time, and the heterogeneous hardware characteristics of end-side equipment are adapted. The method is suitable for an end-side environment with an immutable operating system as a base, and is compatible with a cloud containerization deployment scene.

Methods and systems are presented for providing a framework for facilitating multi-party computation within a sharding environment. After a blockchain is divided into multiple shard chains, a multi-party computation system obtains attributes associated with a first shard chain. The attributes may represent characteristics of the first shard chain, characteristics of transactions recorded in the first shard chain, and characteristics of the computer nodes configured to manage the first shard chain. Based on the attributes, the multi-party computation system determines a multi-party computation scheme that specifies a minimum threshold number of nodes required to participate in a transaction validation process and at least one required node required to participate in the transaction validation process for the first shard chain. The multi-party computation system configures the computer nodes configured to manage the first shard chain to perform the transaction validation process according to the multi-party computation scheme.

The invention provides an optimization method and system for improving the file real-time access speed of an electric power embedded operating system, and belongs to the technical field of electric power embedded systems. The method comprises the following steps: establishing a dynamic preloading mechanism based on the operation state of the power equipment; the dynamic preloading mechanism triggers file preloading operation by monitoring the working condition change of the equipment; designing an improved cache replacement algorithm; the improved cache replacement algorithm is used for dynamically adjusting a cache strategy in combination with power file access characteristics; constructing a multi-dimensional file access scheduling system; the multi-dimensional file access scheduling system allocates system resources according to task priorities; executing a file access full-process acceleration strategy; the file access full-process acceleration strategy comprises metadata management, data prefetching and version control. The file access delay can be shortened, and the system resource utilization rate can be improved.

The invention discloses a method and equipment for testing an AUTOSAR AP application program. According to the method, a calling instruction containing a module, an interface and a parameter data type is obtained by analyzing an ARXML file, a corresponding function template is matched according to the module and the interface in the instruction, the parameter data type is written into the template to generate adaptive stub functions, and finally unit testing of the application program is completed based on the stub functions. According to the method, automation of stub function generation is achieved, manual writing and adjustment are not needed, and the error rate and workload caused by manual operation are greatly reduced; through accurate matching of the parameter data type and the function template, the consistency of the stub function and the application program interface is ensured, and the accuracy of unit testing is improved; the whole process adapts to the standardization characteristic of the AUTOSAR AP, and the efficiency and reliability of application program unit testing are remarkably improved.

The invention discloses a large model reasoning optimization method for SLO perception in an edge heterogeneous computing power network, aiming at an edge heterogeneous computing power computing cluster scene providing large model reasoning service, requests are strategically scheduled to make full use of heterogeneous computing power of a cluster, so that SLO heterogeneous requests are met to the maximum extent; the method comprises the following steps: firstly, separating a pre-filling stage and a decoding stage of a reasoning process to different nodes, and respectively adopting a data parallel deployment strategy and an assembly line parallel deployment strategy to ensure that the computing power of equipment is fully utilized; then, according to heterogeneous computing power characteristics of decoding nodes, a distributed assembly line non-uniform deployment strategy is realized to improve the model reasoning throughput; finally, round rewards are collected in offline exploration, a large model is used for assisting in encoding of potential rewards, a decoder is trained to decode agency rewards in each step, a reinforcement learning PPO algorithm is assisted in training a scheduler, effective scheduling requests are achieved, the resource utilization rate is increased, meanwhile, cluster energy consumption is reduced, and the throughput of large model reasoning service is increased.

The invention provides a hardware description language code generation system based on TypeScript, comprising a configuration conversion unit, a type verification unit, a parameter calculation unit, an intermediate representation construction unit and a code generation unit, through a type system of TypeScript, type inconformity can be effectively captured in a coding stage, illegal configuration with parameter type inconformity is effectively intercepted, and the code generation efficiency is improved. Therefore, the design error risk and the later debugging cost are reduced; meanwhile, based on the object-oriented characteristic of the TypeScript, the target hardware description language code with the structured hierarchy can be generated, so that the readability of the target hardware description language code is high, and then the maintainability and development efficiency of hardware design are improved.

The invention discloses a simulation test method for electronic packaging equipment, which comprises the following steps of: importing PCB (Printed Circuit Board) design data to form a unified intermediate format; multi-dimensional parameter mapping association is carried out on the geometric entity and an equipment execution module of packaging equipment in the digital twin model library, and multi-dimensional parameters comprise space coordinates, thermal characteristics and electrical characteristics; according to the correlation result, a dynamic rule engine is called to generate and update process rules in real time, and the process rules comprise the solder paste jet printing speed, the dispensing pressure, the surface mounting force threshold value and the backflow peak temperature curve; loading the updated process rule to a digital twin model of the electronic packaging equipment, and driving the model to generate an executable control program; and running the control program in a 3D simulation environment to obtain a virtual running track, a heat distribution field and a stress field. According to the method, the trajectory, the thermal field and the stress field are output in advance through 3D simulation, the collision, warping and overheating risks are eliminated in the virtual stage, the number of real object proofing rounds is reduced, and the overall introduction cost is saved.