29 results about "Token bucket" patented technology

The application discloses a kind of electronic bank dynamic flow control method and system based on multidimensional data fusion, belong to data processing technical field.The method includes: the multiple-source data in electronic bank system is collected and unified structured processing is carried out;Dynamic baseline is generated based on historical transaction data using sliding window algorithm, and dynamic baseline is corrected in combination with real-time cluster load state;Business priority and multi-tenant priority are calculated, and after dynamically adjusting weight coefficient according to load factor, weighted summation is obtained to obtain comprehensive priority;According to the comprehensive priority, generate token bucket parameters, execute service level throttling and interface level throttling, and dispatch requests according to queue weight proportion;The execution result is collected and returned to update dynamic baseline and calculation parameters.The application realizes adaptive adjustment and differentiated resource guarantee of throttling parameters through the synergistic effect of dynamic baseline calculation, comprehensive priority model, hierarchical throttling execution and closed-loop feedback.

This application relates to the field of database operation and maintenance technology, specifically to an automated operation and maintenance method for audit system databases. This method includes: periodically collecting inventory levels and thresholds at each level of an LSM-Tree, and calculating the level capacity occupancy ratio; using eBPF technology to statistically analyze background I / O latency and traffic within a sliding window, and calculating the unit latency for background disk writes; performing recursive calculations in reverse order using the level capacity occupancy ratio to obtain the end-to-end logical write multiple; analyzing the coupling relationship between the level capacity occupancy ratios of adjacent levels, determining the multi-level concurrent congestion coefficient, obtaining the estimated unit data occupancy time, and inversely deriving the allowed front-end write rate; and combining this with a token bucket algorithm to smoothly limit the write requests of the front-end audit system. The goal of this application is to achieve early detection of backlog trends within the LSM-Tree and to automatically perform predictive traffic control to optimize the operation and maintenance management of the audit system database.

The application provides a service data processing request flow limiting method, device, equipment and medium. In the method, the service node equipment sends the obtained interface identifier and flow limiting data to the database equipment before sending the service data processing request each time. The database equipment obtains the sending success number, the remaining token number and the called interface moment according to the interface identifier, and further calculates the current token number in the token bucket. When the consumed number of the calling interface token is greater than the current token number, a flow limiting result indicating that flow limiting is needed is generated and sent to the service node equipment. The service node equipment performs flow limiting processing on the service data processing request. According to the present scheme, the database equipment determines whether flow limiting is needed according to whether there are enough tokens in the token bucket. The token bucket capacity number is related to the processing concurrent data number of the service data processing system, which guarantees the processing concurrent data capability of the service data processing system and effectively improves the security of the service data processing system.

The present application relates to the technical field of building robot operation management and job scheduling, and particularly relates to a method and system for building robot task allocation combined with power management, obtaining parameter information, driving information and residual power of a driving section, obtaining a task set to be executed and determining a preset task by calculating and predicting power consumption of a power consumption prediction model; updating the predicted power consumption within a preset time window, determining an entry threshold value / exit threshold value according to a power feasibility judgment sequence to obtain an equivalent flip frequency; setting a task reassignment token bucket replenishment rate and an upper limit according to an invalid energy consumption proportion sequence; allocating and monitoring, reassigning and recalculating if the exit threshold value is lower and the tokens are sufficient, otherwise issuing a charging task, updating energy consumption prediction in real time and selecting the optimal task to reduce misallocation; based on the feasibility flip adaptive entry / exit threshold value, relieving jitter, improving continuity and safety; and feeding back the token bucket according to invalid energy consumption to suppress frequent reassignment, reduce energy consumption and increase efficiency.

The application discloses an Ethernet and serial communication conversion device and method for semiconductor equipment, which comprises a hardware system and a software system, the hardware system contains an industrial-grade main controller, a plurality of independent configurable serial communication interfaces, an Ethernet interface, a state indication and alarm module and the like; the software system runs on the main controller and comprises a bidirectional protocol conversion module, a data priority processing and flow control module and a state monitoring and fault diagnosis module. The data priority processing and flow control module is internally provided with a data processing mechanism based on static priority, or adopts an adaptive priority dynamic re-marking and flow shaping method based on process stage sensing and queue pressure prediction, and the priority of a data frame and a token bucket injection rate are dynamically adjusted. The application can adapt to the harsh environment of a semiconductor factory, realize high-reliability and low-delay bidirectional conversion of a multi-protocol serial device and an Ethernet, and has perfect state monitoring and fault automatic recovery functions.

The application discloses an interface flow limiting method based on a token bucket algorithm. The method collects system load indicators and historical request flow data, dynamically generates a rate adjustment signal through a multi-dimensional weighted scoring mechanism, updates the token generation rate of a local token bucket, and broadcasts the rate parameter to other service nodes under the same interface via a distributed coordination node. In the token generation link, the inventory tokens are sequentially executed for time-limited removal and water level truncation operation. When the request arrives, the token acquisition is atomically executed, and a release signal or a flow limiting signal is generated. The differentiated degradation disposal is performed on the flow limiting request according to the priority of the calling party.

The present disclosure provides an EHPC cluster deployment-based interruption recovery method and system. The EHPC cluster deployment-based interruption recovery method of an embodiment comprises: based on tenant identification of a plurality of deployment tasks, aggregating deployment tasks belonging to the same tenant into batch task groups, and performing service priority classification on each deployment task within the batch task groups; based on the grouped deployment tasks, constructing a layered token bucket of a resource and interface two-dimensional four-layer structure and dividing platform resources used for executing the deployment tasks into mutually isolated reserved core resource pools and shared resource pools; in response to detecting a deployment service restart, determining the admission qualification of the reserved core resource pools according to the service priority within the corresponding batch task group and allocating tokens of the layered token bucket; and performing an interruption recovery process of the deployment tasks allocated through the tokens. The present disclosure embodiment can ensure normal interruption recovery of core tasks under an extreme concurrent scenario.

This invention relates to the field of distributed flow control technology and discloses a dual-channel elastic quota borrowing method based on token buckets and global locks. The method includes: 1) routing client-initiated business requests to a high-concurrency channel or a low-concurrency channel according to preset rules; 2) when a business request is routed to the high-concurrency channel, obtaining a token from the dedicated token bucket of the high-concurrency channel, prioritizing the use of the high-concurrency channel's own quota, and when the high-concurrency channel's own quota is exhausted, performing quota probing and borrowing the quota of the low-concurrency channel when it is idle, allowing the request to pass flow control; 3) when a business request is routed to the low-concurrency channel, firstly constructing a low-concurrency channel occupancy lock, then obtaining a token from the dedicated token bucket of the low-concurrency channel, and allowing the request to pass flow control when the low-concurrency channel has sufficient own quota. Using this invention, high-concurrency channels can safely borrow the quota of low-concurrency channels when they are idle, improving overall throughput.

The application provides a multi-operator short message channel dynamic selection and load balancing method and system, relates to the technical field of mobile communication, and collects the forwarding time consumption of each short message channel at the transmission nodes of the corresponding branch path of the global network topology, the link hop number of the adjacent transmission nodes, and the connection state and switching number of the target user in the corresponding physical area network; based on the forwarding time consumption and the link hop number, the resistance coefficient of each transmission node is obtained; the transmission node is taken as a graph node, the physical link is taken as a directed edge, the delay gradient and the resistance coefficient are taken to determine the edge weight, the upper limit of the physical bandwidth of the short message channel is taken as the capacity constraint, a weighted directed graph is constructed; in the weighted directed graph, a capacity allocation matrix is constructed; a token bucket is set for each short message channel, and the dynamic verification and distribution of the concurrent short message sending request are carried out according to the available capacity, so that the accurate congestion avoidance and efficient load balancing of the multi-operator short message channel in the complex network environment are realized.

The application provides an API construction method of industrial big data microservice, and belongs to the technical field of industrial big data.The application adopts Maxwell daemon control mechanism to calculate the information value score of API request by establishing a distributed API gateway cluster, and realizes hierarchical queue scheduling by combining a token bucket algorithm, deploys a service grid sidecar agent and a kernel bypass module based on eBPF to generate an optimal calling path by predicting the service node response delay by using a time sequence dependency graph analysis model, and adopts a hybrid logical clock protocol to perform differential consistency processing on control type API requests and monitoring type API requests, and builds an LSM tree storage engine to cooperate with a time partition strategy and a Z-Order curve multidimensional index to persist data, so that the problem that key control business is blocked by ordinary monitoring requests due to the inability to effectively distinguish different value API requests is solved.

This application discloses a traffic shaping method, a traffic shaping device, a video device, and a computer program product. The method includes: classifying traffic within the video device based on preset traffic characteristics; allocating corresponding token buckets to each type of traffic and setting shaping parameters for each token bucket based on a preset shaping strategy; determining a target parameter adjustment strategy from a preset first parameter adjustment strategy and a second parameter adjustment strategy based on the bandwidth status of the video device, wherein the first parameter adjustment strategy and the second parameter adjustment strategy impose different constraints on the shaping parameters; adjusting the shaping parameters of each token bucket according to the target parameter adjustment strategy; and shaping and transmitting the corresponding type of traffic based on the latest shaping parameters of each token bucket. This application's solution can improve the transmission performance of various types of traffic by video devices under dynamic bandwidth conditions.

This application provides a QoS control method and system based on IO demand prediction. The method includes: capturing the IO system call trajectory of a target AI task and extracting an enhanced IO feature vector, where the enhanced IO feature vector is a feature vector containing system call features and file metadata features; matching a pre-trained enhanced IO fingerprint library to identify the current execution stage; predicting the IO demand of the next stage based on a Hidden Markov Model (HMM), where HMM is a probabilistic model used for sequence state prediction; and directly calling the kernel's native block device layer QoS interface to update the token bucket parameters of the target AI task based on the prediction results, where the block device layer QoS interface refers to the token bucket parameter update interface provided by the Linux kernel's rq-qos module. This application achieves non-intrusive prediction of AI task IO demands and ultra-low latency dynamic control of block device layer bandwidth without modifying application code or kernel code.

The application relates to the technical field of multi-agent collaboration and automation process, in particular to an Agent and RPA asynchronous conflict solving and scheduling framework system and method for complex tasks, which comprises a task planning module, a conflict management module, an environment perception module, an abnormality self-healing module and a resource monitoring module. The system uses a large language model to decompose a task into atomic flows, realizes resource asynchronous scheduling and race control by constructing a dynamic conflict matrix and introducing a token bucket mechanism, and simultaneously monitors an environment state in real time and captures an execution snapshot. When detecting an environment exception or interruption, self-healing logic is triggered based on environment consistency scoring, a breakpoint is located by using the snapshot, and incremental recovery is carried out. The application can effectively solve the conflict problem of multiple agents for exclusive resources, improve the robustness of RPA in a complex environment, significantly reduce the cost of manual intervention, and ensure the timely response of high-priority tasks.

The present application relates to a rate limiter supporting a prioritized batch traffic request. The rate limiter comprises: a traffic splitting module, used for splitting a batch traffic request into a plurality of batch fragments; a token management module, used for recording and managing a current token quota accumulated in a token bucket; and a rate limiting determination and execution module, used for performing rate limiting determination on the batch traffic request on the basis of a rate limiter parameter and the current token quota, and executing, on the basis of a determination result and by means of the traffic splitting module, an action of allowing or splitting the batch traffic request.

The application provides a multi-source adaptive enterprise micro group robot message intelligent pushing method and device, and relates to the technical field of computer application. The method realizes the format unification and effectiveness guarantee of different source business data through unified access and standardized preprocessing of multi-source business data, and realizes professional processing of massive real-time raw data through a distributed stream processing engine. Through the combination of multiple trigger modes and distributed lock control, reliable triggering of full-scene pushing tasks is realized, and the problem of repeated execution of tasks in a distributed cluster environment is completely solved. Through standardized content rendering and traffic control based on the token bucket algorithm, accurate adaptation of pushing content and enterprise micro robot API is realized, avoiding API ban problems in high-concurrency scenarios, and comprehensively improving the efficiency, success rate, stability and scene adaptation ability of enterprise micro group message pushing, which can meet the enterprise micro message pushing needs in multiple scenarios of the financial, Internet and other industries.

The application discloses a multi-type physical interface conversion method based on software definition, S1: an interface definition file is acquired, and a port configuration set is generated by analysis according to port granularity; S2: a port reconstruction vector is generated based on the port configuration set, port software reconstruction is carried out, and automatic rollback is performed when an abnormal configuration occurs; S3: a unified timestamp is marked to multi-port input data, and the multi-port input data is encapsulated into a standardized input unit; S4: a basic protocol plug-in is loaded, and protocol analysis is carried out, and a structured message is output; S5: multi-port data alignment is carried out, and fixed priority and a token bucket are adopted to perform deterministic scheduling and congestion control; and S6: data encapsulation and differential output are carried out. The port access elements are uniformly described by the interface definition file, and a port reconstruction vector is automatically generated to realize hot switching and rollback, so that the same device and the same method can be reused by only replacing the interface definition file in different projects, and when wiring differences occur, the wiring can be corrected through configuration, and manual wiring change and downtime are reduced.

The application discloses a method for solving message accumulation, and relates to the field of message sending management; the method comprises the following steps: step 1, configuring parameter information in a message processing flow; step 2, monitoring the production speed of different types of messages, the consumption speed of the messages, the resource usage of the message consumers, and the response situation after the messages are consumed; step 3, performing message flow control according to the message monitoring situation, limiting the message receiving and sending frequency, and specifically adopting a token bucket mode to perform the message flow control; and step 4, processing message accumulation: referring to a message accumulation ratio to obtain the accumulation situation of the message queue at this time, continuously reducing the generation quantity of tokens in a unit time for a message receiving token bucket of a message type that is in an accumulation state when the message queue is about to be filled; performing a retrieval and deduplication operation on the messages that have not been forwarded; performing disk storage on the accumulated messages, and emptying the token quantity of the message receiving token bucket of the accumulated messages, so that the messages of this type are no longer received.

This application discloses a flow control method, apparatus, device, and medium for input / output requests, relating to the field of computer technology. The method includes: responding to an input / output request corresponding to a disk array, obtaining the request type of the input / output request, wherein the request type is an access request or a maintenance request, where an access request is used to enable an external device to perform data access operations on the disk array, and a maintenance request is used to enable internal maintenance operations on the disk array; if the request type is an access request, determining the request queue corresponding to the input / output request as a token bucket queue; or, if the request type is a maintenance request, determining the request queue corresponding to the input / output request as a leaky bucket queue; and adding the input / output request to the request queue corresponding to the input / output request. This application combines the token bucket and leaky bucket mechanisms, balancing the burstiness of access requests with the stability of maintenance requests.

The application relates to the technical field of gateway acceleration, and discloses a gateway-end intelligent acceleration system and method, which comprises the following steps: selecting multiple edge nodes according to the geographical position coordinates of a gateway and establishing a pre-connection edge node channel; matching in-coming data packets of the gateway with a game feature library to obtain a local traffic recognition result and generate game traffic marked with a priority level together with application type labels returned by an operator DPI device; calculating an end-to-end forwarding path from the edge nodes of the pre-connection edge node channel, encapsulating the path into a routing instruction message, and delivering the message to the gateway; and creating a layered token bucket queue in the gateway and forwarding the game traffic marked with the priority level according to the end-to-end forwarding path specified by the routing instruction message. The application establishes a pre-connection channel mechanism between the gateway and the edge nodes of the operator, so that the connection does not need to be re-established when a game is started, the delay of cross-region server switching is reduced, and the smoothness of the user game experience is improved.

The present application relates to the technical field of flow control, in particular to a dynamic advertisement frequency control optimization method based on multi-dimensional evaluation of family members, comprising obtaining device topology features and interaction logs to construct a heterogeneous mapping graph, and identifying a family unit by using graph representation learning and community discovery algorithm; by extracting behavior characteristics such as key Shannon entropy, time interval and response jitter, a classification model is used to determine and activate the active member identity; based on the user activity index and historical load data, a dynamic token bucket is initialized, and the load pressure coefficient is calculated by combining resource saturation and global residual water level weighting; the depth of the token bucket is dynamically adjusted by using a nonlinear negative feedback mapping mechanism to adjust the upper limit of frequency control in real time; cooperate with the flow interception hook and the double atomicity check mechanism to execute data distribution decision. Through identity perception and negative feedback regulation, the present application realizes adaptive optimization of quota allocation.

The application provides a block device layer differentiated admission control method and system, the method comprises: obtaining input and output request characteristic data of the block device layer, the input and output request characteristic data is obtained by feature extraction on the input and output request; identifying the key value cache access mode according to the input and output request characteristic data; according to the identified key value cache access mode, the corresponding token bucket resource is allocated to the input and output request, and the differentiated admission control is executed. Without modifying the upper reasoning engine, the application realizes the differentiated block device layer admission control of the key value cache flow and the conventional data flow in the large model reasoning, and reduces the delay of the key value cache reading.

The application relates to the field of data processing and discloses a user concurrent access control method and related device, which are used for improving the efficiency of user concurrent access. The method comprises the following steps: obtaining the access authority and priority of a plurality of target users, and performing token distribution on the plurality of target users according to the access authority and priority, so as to obtain the token quantity; obtaining at least one first token according to the token quantity corresponding to each target user, and caching the at least one first token to a preset token bucket, and setting the maximum capacity of the token bucket; when a target user accesses a shared resource, obtaining a target quantity of at least one second token from the token bucket; when the target user completes the access to the shared resource, releasing the at least one second token, and putting the at least one second token back into the token bucket; periodically checking the token quantity in the token bucket, and dynamically adjusting the token quantity and the acquisition rate of the token bucket according to the access state.

This disclosure provides a traffic shaping method, apparatus, medium, and electronic terminal. The traffic shaping method includes: acquiring a queue request signal; allocating a counter corresponding to the traffic shaping mode of the queue within the port of the vehicle system to the queue, the counter being a credit counter or a token counter, the counter supporting deficits and capable of storing several maximum packet lengths supported by the vehicle system; and performing traffic shaping on the request queue associated with the queue request signal based on the counter, the credit increment or token bucket increment of the request queue being increased once every fixed time interval via a flag signal. In this traffic shaping method, performing traffic shaping based on the counter improves traffic shaping accuracy. Furthermore, the vehicle system supports both credit-based and token-based traffic shaping methods, making traffic shaping in the vehicle system more flexible.

A system that provides services to clients may receive and service requests, various ones of which may require different amounts of work. The system may determine whether it is operating in an overloaded or underloaded state based on a current work throughput rate, a target work throughput rate, a maximum request rate, or an actual request rate, and may dynamically adjust the maximum request rate in response. For example, if the maximum request rate is being exceeded, the maximum request rate may be raised or lowered, dependent on the current work throughput rate. If the target or committed work throughput rate is being exceeded, but the maximum request rate is not being exceeded, a lower maximum request rate may be proposed. Adjustments to the maximum request rate may be made using multiple incremental adjustments. Service request tokens may be added to a leaky token bucket at the maximum request rate.

Various embodiments provide for time-based token bucket for controlling data rate, which can be used by a network device (e.g., multi-layered network device). An embodiment described herein can emulate accumulation of number of tokens (e.g., in a token bucket) using precision timestamps.

This application provides a block device layer differentiated admission control method and system. The method includes: acquiring input / output request feature data of the block device layer, wherein the input / output request feature data is obtained by feature extraction of input / output requests; identifying key-value cache access patterns based on the input / output request feature data; allocating corresponding token bucket resources to input / output requests according to the identified key-value cache access patterns; and performing differentiated admission control. This application achieves differentiated block device layer admission control between key-value cache traffic and regular data flow in large model inference without modifying the upper-layer inference engine, reducing the latency of key-value cache reads.

The present application relates to the technical field of computers, and in particular to a request processing method and apparatus, an electronic device, and a storage medium, for use in preventing hallucinations in large models due to high concurrency, and improving the availability of the large models. The method comprises: acquiring a request to be processed, initially acquiring from a token bucket a token corresponding to said request, and adding 1 to the total number of concurrencies of a current sliding time window; if the initial acquisition of the token fails, on the basis of the total number of concurrencies, and the throughput, average delay and maximum acceptable delay of a large model, determining the number of retries and a retry interval duration of re-acquiring the token; and on the basis of the number of retries and the retry interval duration, re-acquiring the token from the token bucket, and when the re-acquisition of the token succeeds, performing inference on said request on the basis of the large model, and generating and returning an answer. In this way, the number of retries and the retry interval duration are dynamically adjusted on the basis of the real-time total number of concurrencies, and requests are scattered to reach large models, thereby improving the response success rate and availability.

The present disclosure provides an EHPC cluster deployment-based interruption recovery method and system. The EHPC cluster deployment-based interruption recovery method of an embodiment comprises: based on tenant identification of a plurality of deployment tasks, aggregating deployment tasks belonging to the same tenant into batch task groups, and performing service priority classification on each deployment task within the batch task groups; based on the grouped deployment tasks, constructing a layered token bucket of a resource and interface two-dimensional four-layer structure and dividing platform resources used for executing the deployment tasks into mutually isolated reserved core resource pools and shared resource pools; in response to detecting a deployment service restart, determining the admission qualification of the reserved core resource pools according to the service priority within the corresponding batch task group and allocating tokens of the layered token bucket; and performing an interruption recovery process of the deployment tasks allocated through the tokens. The present disclosure embodiment can ensure normal interruption recovery of core tasks under an extreme concurrent scenario.