214 results about "Computer resources" patented technology

The invention discloses an intelligent resource scheduling system and method based on an elastic threshold value and AI prediction, and relates to the technical field of computer resource management. For the limitation of the existing resource scheduling method, the provided scheme comprises an elastic threshold configuration module used for defining performance indexes of computing resources, allocating weights and setting initial upper and lower limits of an elastic threshold; the dynamic threshold value generation module is used for collecting and calculating resource performance index data in real time, calculating a real-time dynamic threshold value after preprocessing, and comparing the real-time dynamic threshold value with an elastic threshold value range; the AI prediction module is used for training and optimizing a prediction model, and the model obtains a future computing resource demand prediction result based on the new data; and the real-time scheduling engine module is used for formulating a resource scheduling strategy, distributing computing resources, executing computing resource increasing and decreasing operation, monitoring a scheduling result and feeding back the scheduling result, so that the modules are adjusted, and a dynamic optimization closed loop of the resource scheduling strategy is formed. The method is used for improving the computing resource utilization rate.

Systems and methods for a multi-camera object detector having two or more backbone models. In particular, systems and methods are provided for including two or more backbone machine learning models, with one backbone optimized for speed and the other backbone optimized for precision. In particular, the first backbone can be highly accurate but slower than the second backbone and with a higher computer resource usage. The second backbone can be fast and efficient but have lower accuracy for object detection. In some examples, the second backbone can use fewer images and / or lower resolution images. The determination of which backbone to use can be based on fixed rules, or it can be determined based on another machine learning component. The outputs from the first and second backbones for each camera can be combined together into a unified representation, such as a bird's eye view (BEV) space.

This invention discloses a full lifecycle management system and method for scientific computing programs. The system includes a build environment subsystem and a production environment subsystem. The former provides computer resources for the build process of the scientific computing program throughout its lifecycle, while the latter provides computer resources for the testing and deployment processes. This invention, through a full lifecycle management method for scientific computing programs, operates on corresponding computing resources, encompassing a series of steps including querying, triggering scheduling, build execution, result distribution, and test deployment. It automatically generates the executable file of the scientific computing program and configures its runtime dependencies. Simultaneously, it automatically generates a corresponding description file recording the entire lifecycle process. Furthermore, based on version management of these description files and their sets, it achieves full lifecycle traceability and cross-platform migration and deployment of scientific computing programs in a high-performance computing environment.

The invention belongs to the field of computer resource management, particularly relates to a computer resource dynamic allocation management method based on cloud computing, and aims to solve the problems of locality of a decision view angle and hysteresis of a time dimension in related technologies. The method comprises the following steps: acquiring the resource utilization rate of each virtual machine in a cloud environment, wherein the resource utilization rate comprises a historical resource utilization rate and a real-time resource utilization rate; according to the resource utilization rate, predicting a resource demand of each virtual machine in a future preset time period; determining a resource competition conflict according to the predicted resource demand, and constructing a resource competition graph model; mapping the resource competition graph model into a non-cooperative game model; and determining a scheduling scheme in the cloud environment based on the non-cooperative game model, and executing a resource allocation operation according to the scheduling scheme. According to the method, potential conflicts can be identified and scheduled before resource competition actually occurs, and performance reduction caused by insufficient resources is reduced.

Computer-implemented methods for recommendation prioritization for a container orchestration system. Aspects include receiving a set of recommendations for a cluster of a container orchestration system. Aspects also include selecting an optimal recommendation from the set of recommendations using a scored knowledge transform graph. Aspects further include generating a confidence score for the cluster based on the optimal recommendation. Aspects also include determining a category of a readiness assessment model for the cluster using the confidence score. Aspects further include modifying a computer resource of the cluster based on the category of the readiness assessment model.

A method of causing ransomware to be executed includes obtaining a set of ransomware samples and automatically performing a set of scripted detonations of a first subset of the ransomware samples according to a schedule. The set of scripted detonations includes executing the first subset of ransomware samples on a virtual machine. During the execution of the first subset of ransomware samples, use of computing resources by the first subset of ransomware samples is monitored until a shutdown threshold is reached. Upon reaching the shutdown threshold, any pending detonation tasks of the execution of the first subset of ransomware samples are purged. Detonation-infected disk images are generated to include files resulting from the execution of the first subset of ransomware samples, and the detonation-infected disk images are used in training a machine learning system to assess a likelihood of a presence of a ransomware attack.

The invention discloses an RAG-oriented embedded service flexible deployment method, and belongs to the technical field of computer resource configuration. The method comprises the following steps: firstly, obtaining an embedding model suitable for an RAG scene, constructing an embedding service, testing different resource configuration parameters in batches through automatic resource analysis, and screening out optimal resource configuration meeting service quality requirements; secondly, an RPS monitoring and dynamic batch processing mechanism is integrated, and a GPU resource fragmentation optimization algorithm and a three-level candidate GPU scheduling strategy are combined; triggering a hybrid capacity expansion and contraction mechanism according to a reasoning request load to form cooperative scheduling; and finally, reasoning requests are distributed through a polling mechanism, and service performance is continuously optimized. Through the synergistic effect of automatic resource analysis, GPU resource fragmentation optimization and hybrid capacity expansion and contraction strategies, the GPU resource utilization rate and embedded service performance can be remarkably improved, and the high-concurrency and low-delay reasoning requirements in an RAG scene are effectively met.

The invention discloses a computer resource scheduling optimization method based on artificial intelligence, and relates to the technical field of computer resource allocation, and the method comprises the steps: constructing a twinborn computer space, marking twinborn collection ends, and collecting resource use data; constructing an original coordinate system according to the resource use data and performing state generation to obtain a resource use state diagram; performing traversal check on the resource use state diagram to obtain a periodic fluctuation segment, and performing resource quantification through a twin computer space to obtain a fluctuation proportion accommodating frame; monitoring and deploying the CPU based on the twinborn computing space to obtain virtual deployment resources, applying the virtual deployment resources to the computer, and generating an optimal scheduling strategy according to the application process; resource allocation is optimized, the CPU utilization rate is improved, computer operation pressure is reduced, and idle or overuse of resources is avoided.

The invention discloses a probabilistic route map path planning method and device. The method comprises the following steps: S1, acquiring global position information and obstacle information of an automatic driving vehicle; s2, sampling to obtain candidate path nodes; s3, constructing a covariance matrix; s4, calculating the mahalanobis distance between each candidate path node and the barrier distribution; s5, discarding sampling points whose mahalanobis distances are smaller than a set safety threshold, and performing preliminary screening on candidate path nodes; s6, constructing a probability route map based on the screened candidate path nodes; and S7, completing path planning. According to the method, useless and low-efficiency sampling nodes can be accurately eliminated, so that the construction of the probabilistic route map is simpler and more efficient, and the construction quality and effect of the probabilistic route map are greatly improved; meanwhile, interference and calculation burden caused by invalid nodes are reduced, and the efficiency of path planning is remarkably improved; in addition, waste of computer resources is reduced, and resource utilization efficiency is improved.

The invention discloses an online resource allocation method and system based on multiplicative weight updating, and belongs to the technical field of computer resource scheduling. In order to solve the problems that the resource utilization rate is low, the service quality is difficult to guarantee and the feedback information is limited in a multi-user sharing scene, a multiplicative weight updating strategy based on a binary queue state is mainly adopted, and the resource allocation proportion is dynamically adjusted in combination with a truncated simplex projection mechanism. According to the method, collaborative optimization of resource utilization efficiency maximization and service quality guarantee can be realized under the condition of only depending on the empty / non-empty state of the queue, and the method has good efficiency, robustness and fairness.

The invention discloses a graphics card multi-task co-processing system based on dynamic video memory allocation, which belongs to the technical field of computer resource management and comprises a task grading module, a video memory state scanning module, an allocation strategy module and a resource recovery integration module. The task grading module is used for acquiring tasks, dividing the tasks into high-priority tasks and low-priority tasks, and establishing an independent resource pool according to initial priorities of the tasks; the video memory state scanning module is used for scanning a video memory use state in the independent resource pool and generating a video memory resource distribution map; the allocation strategy module is used for generating an allocation instruction based on the video memory resource distribution graph and the task priority; and the resource recycling and integrating module is used for recycling the continuous blocks released after the high-priority task is finished and updating the video memory resource distribution map. Through the task grading module, the video memory state scanning module, the allocation strategy module and the resource recovery integration module, the system response efficiency is improved.

The invention discloses a test and evaluation method for a virtualization system in multiple scenes, which comprises the following steps: deploying an executable file compiled by a test script at a virtualization client, and testing the performance of a client operating system to judge the performance loss of the virtualization system to computer resources. The method comprises the following steps: step 1, creating a project, and configuring system parameters; 2, integrating a test program into a project, and generating a mirror image file; step 3, packaging a complete virtualization system mirror image; 4, transmitting the mirror image to a development board through a network protocol; 5, executing a command to start a system, and logging in a system console through a serial port; and 6, calling and executing the function test script, the performance test script and the security test script.

The invention provides a display method and device, electronic equipment and a computer medium, and the method comprises the steps: displaying a preset scene picture which comprises a plurality of objects; in response to detection of a moving operation for a first object in the multiple objects, multiple pieces of target bounding box information corresponding to the multiple objects are obtained, and each piece of target bounding box information in the multiple pieces of target bounding box information is used for representing a parameter of an axis alignment bounding box corresponding to the object corresponding to the target bounding box information; based on the multiple pieces of target bounding box information, determining a to-be-adsorbed object corresponding to the first object in the multiple objects; and based on the target distance information between the first object and the to-be-adsorbed object, determining whether to display an adsorption effect of adsorbing the first object to the to-be-adsorbed object. According to the technical scheme provided by the embodiment of the invention, the effects of improving the calculation efficiency and saving computer resources can be achieved.

The invention discloses a computer resource allocation method and system based on big data analysis, and relates to the technical field of computers, and the method comprises the steps: collecting computer resource data in real time, carrying out the preprocessing, and predicting the future resource demands of a computer through a reinforcement learning model based on the preprocessed data; based on future resource requirements, a computer resource allocation strategy is generated by combining a particle swarm optimization algorithm with a cloud model disturbance mechanism and self-adaptive inertia weight adjustment, and a dynamic scheduling mechanism of cloud computing and edge computing is utilized. According to the method, key parameters of the particle swarm optimization algorithm are set and initialized, and adaptive inertia adjustment, cloud model disturbance and an auxiliary particle replacement mechanism are combined, so that the global search capability and diversity control of the particle swarm are enhanced, local optimum is effectively avoided, the quality and convergence stability of a resource allocation solution are improved, and the resource allocation efficiency is improved. Therefore, the accuracy and efficiency of computer resource allocation are remarkably improved.

Sanitizing data can be a cumbersome task, particularly when the volume of data is large, the content is sensitive, and / or the type of sanitation requires contextual determinations. Sanitizing large amounts of data is tedious and may often require highly trained personnel with clearances and / or other qualifications. In the systems and methods of the present disclosure, language models (LMs) are used to solve these and other technical issues with tools that may allow sanitizing data easily, with high versatility, context awareness, and / or low demand for computational resources. In particular, some of the disclosed systems and methods use a first language model and a second language model (being less resource-intensive than the first language model) to generate sanitized output data with improved efficiency and accuracy. This dual-model approach ensures that sensitive information is handled appropriately while optimizing computer resource usage.

A routing circuit for an integrated circuit is configured to access a set of resources organized according to a topology having multiple dimensions. The routing receives a request for a particular resource in the set of resources, the request including an address that includes a first set of bits and a second set of bits, the topology having a first dimension with n routing options (where n is not a power of two) and a second dimension with m routing options. The routing circuit determines a first routing choice and a second routing choice for the first dimension and the second dimension by performing respective modulo-n and div-n operations on a value formed by the address including the first set of bits and the second set of bits. The routing circuit then activates one or more selection signals according to the first routing choice and the second routing choice, the one or more selection signals usable to cause the particular resource to be selected in response to the request.

The invention relates to the technical field of computer resource management, in particular to a server resource scheduling system oriented to a high-density computing environment. The system obtains scheduling data of each dimension in real time, wherein the scheduling data comprises task resource data; according to the difference between the actual value and the predicted value in the scheduling data of the dimension at the current moment and the task resource data, obtaining a resource supply priority degree; obtaining a resource adjustment degree according to the resource supply priority degree and a difference between an actual value and a predicted value of task resource data of a preset specified historical task of the same resource demand type of each task; and based on the task resource data and the resource adjustment degree of each task at the current moment, performing scheduling processing on server resources of the high-density computing environment at the current moment. According to the method, the resource adjustment degree of each task at the current moment is accurately acquired, so that the prediction resource of each task at the current moment can be accurately adjusted, and the efficiency and the stability of task execution are effectively improved.

The invention is notably directed to a computer-implemented method, which aims at jointly identifying an optimal source of computerized resources and optimizing a configuration of the computerized resources. The method comprises configuring a Best-Arm Identification algorithm, in order to (i) associate arms of the algorithm with respective sources of computerized resources and (ii) connect the arms to one or more optimizers. Each of the optimizers is designed to optimize a configuration of such computerized resources. Next, the method iteratively executes the Best-Arm Identification algorithm to progressively eliminate the sources, with a view to eventually identifying one of the sources as an optimal source with an optimized configuration. Several iterations are accordingly performed. During each iteration, each of the arms is pulled and the rewards earned by pulling the arms are computed. Pulling each arm causes to optimize a configuration of computerized resources of a respectively associated source.

The embodiment of the invention provides a cloud computer control method and device, electronic equipment and a storage medium, and the method comprises the steps: determining a cloud computer resource pool in response to a selection operation for resource capacity; in response to a configuration operation for the cloud computer, creating a plurality of different to-be-run cloud computer instances in the cloud computer resource pool, the different to-be-run cloud computer instances being mutually exclusive; under the condition that a cloud computer connection request is received, determining a target cloud computer instance from the to-be-operated cloud computer instances according to the cloud computer connection request; performing concurrent verification based on the target cloud computer instance; and under the condition that the verification is passed, accessing the target cloud computer instance. According to the embodiment of the invention, computing resources of the cloud computer can be efficiently utilized.

The present disclosure relates to an interaction method, device, equipment and storage medium, the method comprising: displaying a first page, the first page being used at least for displaying a connection game screen between a first account object and a second account object; in response to an end instruction of the current round of the game, displaying a first special effect element acting on the second account object; the first special effect element being related to at least identification information of the first account object. The present disclosure improves the propagation efficiency and effect of the interaction form based on the special effect element, and also improves the utilization rate of computer resources.

This application provides a blockchain-based method, device, and medium for vehicle network reputation consensus. The method includes: obtaining vehicle reputation values ​​based on the objective behavior of vehicles in the vehicle network; roadside units collecting vehicle reputation values ​​and temporary reputation values ​​from the vehicle network, calculating a cumulative reputation value based on the vehicle reputation values ​​and temporary reputation values ​​of vehicles participating in message interaction in the current consensus round, and temporarily storing the cumulative reputation value in the roadside units until consensus is reached and block confirmation is performed, then returning the cumulative reputation value to the vehicle nodes; obtaining a list of roadside units and filtering it using a preset filtering mechanism to obtain a leader list; selecting the roadside unit with the largest cumulative reputation value as the leader based on the feedback cumulative reputation value, performing distributed consensus, and broadcasting the message. This application improves the credibility of vehicle network information, uses roadside units as consensus nodes to ensure fairness and efficiency, and reduces computer resources and consensus centralization.

This approach aims to curb the increase in computer resource usage while simultaneously encouraging users to purchase access to paid content. [Solution] The content management system 1 includes a content distribution unit that distributes image content, a thumbnail generation unit that generates thumbnail images corresponding to two points in time in the image content and associates the two points in time with the thumbnail images, an interface that specifies the reference point of the image content, a content display unit that displays a thumbnail image associated with two points in time that include the reference point, and a determination unit that determines whether the user has the right to view the previous image content. The content display unit does not display the image content at a point in time when the user does not have the right to view it, and displays a thumbnail image associated with two points in time that include the reference point when the reference point is a point in time when the user does not have the right to view it.

The application provides an AI intelligent calculation platform reasoning resource scheduling method and system, relates to the technical field of computer resource management and task scheduling, and obtains task load information of a to-be-reasoned task, memory allocation records of an AI intelligent calculation platform in a preset time window, and a retransmission event timing sequence of a bus data link layer; obtains the discrete entropy of each computing node through a Shannon entropy formula according to the memory allocation records; counts the retransmission times of the retransmission event timing sequence in a sliding time window to generate a communication congestion index of each computing node; determines a weight coefficient according to the task load information, constructs a weighted bipartite graph, and determines a matching edge set that satisfies the most matching edges and the smallest sum of edge weights in the weighted bipartite graph by using a Hungarian algorithm, so as to be used for resource scheduling of the to-be-reasoned task. The problem of throughput decline caused by serious memory fragmentation and communication interference in the large model concurrent reasoning scene is solved.

A system includes a memory storing data associated with fuel dispensing operations and a processor operably coupled with the memory. The processor determines that a fuel dispensing operation is anomalous based at least in part on performance of the fuel dispensing operation by determining a measured volume per unit time parameter associated with fuel dispensed from a fuel dispensing terminal and comparing the measured volume per unit time parameter with a threshold volume per unit time parameter. The processor determines that the measured volume per unit time parameter is less than the threshold volume per unit time parameter, wherein the measured volume per unit time parameter being less than the threshold volume per unit time parameter indicates that the fuel dispensing terminal is manipulated to dispense fuel at a flow rate lower than an expected flow rate. In response to determining that the fuel dispensing operation is anomalous, the system deploys computer resources configured to determine a subsequent anomalous fuel dispensing operation before completion of the subsequent anomalous fuel dispensing operation.

Procedure that includes: Received by an authorization server (201) and by a native application (202) on a unit, an initial authorization grant, a public key of a private / public key pair generated on the unit, and an acknowledgment of the authenticity of the native application (202); Received by the authorization server (201) and by the native application (202) on the unit, an update token and a digital signature of the update token created with the private key, wherein the authorization server (201) recognizes the update token only when the update token is verified with the public key that was previously registered; Validate, by the authorization server (201), the digital signature of the update token; In response to the validation of the update token from the native application (202) on the unit, the authorization server (201) transmits a new access token and a new update token from the authorization server (201) to the native application (202) on the unit, the new access token enabling the native application (202) on the unit to access the computer resource; Input into a neural network (757) the initial authorization, the public key, the confirmation of the authenticity of the native application (202), the update token, and the digital signature of the update token, wherein the neural network (757) is trained to generate an access token generated by the neural network (757) and an update token generated by the neural network (757); and Compare, by one or more processors, the new access token and the new update token generated by the authorization server (201) with the access token and the update token generated by the neural network (757) to validate the new access token and the new update token generated by the authorization server (201).

This invention relates to the field of databases, and in particular to a backup method and apparatus based on a distributed graph database. It mainly includes: generating a corresponding transaction accessible table according to a backup command, and broadcasting the transaction accessible table to all graph nodes that need to be backed up; each graph node generating a top-level backup task based on the transaction accessible table, and then deriving backup tasks for each level from the top-level backup task; each graph node asynchronously executing the bottom-level backup task, obtaining user data for backup through the bottom-level backup task, and backing up the user data on its own graph node. This invention can improve the rational utilization of computer resources, achieve the effect of fully parallel backup, and greatly improve the efficiency and performance of backup. It also reduces network overhead during recovery, significantly improving recovery efficiency.

The application discloses a kind of variable scale performance prediction methods of fuel cell system combined with neural network, construct the proton exchange membrane fuel cell model combined with neural network algorithm, realize the variable scale performance prediction of proton exchange membrane fuel cell system, and the proton exchange membrane fuel cell model includes air supply module, hydrogen supply module, water-thermal management module, battery stack module and neural network fitting module;Actual output power of fuel cell system is calculated;After all module combination connection, it is constituted into a variable scale numerical simulation model of proton exchange membrane fuel cell system combined with neural network algorithm.The application combines the advantages of neural network and mathematical modeling, uses limited amount of data resources, computer resources and time cost, completes high-precision simulation modeling of proton exchange membrane fuel cell, to provide guidance for the application of proton exchange membrane fuel cell in aviation field.

The invention relates to the technical field of data processing, in particular to a coding process optimization method and system and a storage medium, and the method comprises the steps: receiving FEC coded audio and video data transmitted by a transmitting end, and obtaining received data; wherein the RTP extension header of the audio and video data subjected to FEC coding carries mark information; determining whether the received data is matched with to-be-transmitted audio and video data corresponding to the FEC coded audio and video data based on the mark information; and if the received data is matched with to-be-transmitted audio and video data corresponding to the FEC coded audio and video data, sending the to-be-transmitted audio and video data to an audio and video decoder for decoding. Thus, if the receiving end completely receives the coded data sent by the sending end, FEC decoding is directly skipped, the audio and video data to be transmitted are sent to the audio and video decoder to be decoded, the step of FEC decoding is not needed, computer resources are saved, FEC processing time is shortened, and waste of computing resources is avoided.

The embodiment of the invention provides a data processing method, a model training method and related equipment, an artificial intelligence technology can be used for processing multi-modal data, the method comprises the following steps: obtaining first data, performing feature extraction on an image in the first data to obtain first feature information of the image, and according to the first feature information, obtaining a model of the image; weight parameters of a first neural network layer in the machine learning model are updated to obtain an updated machine learning model, the updated machine learning model comprises the updated weight parameters of the first neural network layer, and input of the machine learning model comprises texts and / or voices in the first data; obtaining prediction information which is generated by the machine learning model and corresponds to the first data; according to the invention, a scheme capable of processing the multi-modal data is provided, and the length of the text and / or voice input into the machine learning model is not changed, so that the consumption of extra computer resources caused by processing the multi-modal data is reduced as much as possible.

This invention proposes a service creation method, system, and electronic device for a Kubernetes system. The method includes: receiving a request to create a target service in the Kubernetes system; predicting the computer resource usage of the target service based on resource information of already running services related to the target service in the Kubernetes system; creating a container for the target service in the Kubernetes system and configuring initial computer resources for the container based on the predicted computer resource usage; running the target service on the container with the initial computer resource configuration and obtaining the container's computer resource usage; and adjusting the initial computer resources configured for the container based on the obtained computer resource usage, with the adjusted container used to run the target service. This invention can automatically configure a running container with appropriate computer resources for newly created services in the Kubernetes system. Because it does not rely heavily on manual intervention, it can effectively improve the convenience and ease of use of creating services in the Kubernetes system.