30 results about "Throughput (business)" patented technology

An automated system and method is provided for system architects to model enterprise-wide architectures of information systems. From an initial model of a proposed system architecture, performance metrics are modeled and compared against a set of user-defined corporate and business requirements, including cost, quality of service and throughput. For unacceptable metrics, modifications to the system architecture are determined and proposed to the system architect. If accepted, the model of the system architecture is automatically modified and modeled again. Once the modeled performance metrics satisfy the corporate and business requirements, a detailed description of the system architecture derived from the model is output. The model of the system architecture also enables a business ephermeris or precalculated table cross referencing enterprise situation and remedy to be formed. A rules engine employs the business ephemeris and provides indications to the enterprise user for optimizing or modifying components of the enterprise system architecture.

The invention discloses an SLA-based stateless cloud workflow load balancing scheduling method. According to the requirements of different tenants on the business scene and the flow model, different SLA levels are selected, different SLA levels are adopted; the cloud workflow system provides different request throughput services for the tenants; grading service is carried out on different processrequests; engine load real-time monitoring and process model distribution condition on an engine are realized by combining a shared memory; the overall memory overhead of the engine cluster is reducedwhile the request wave crest of the engine service is reduced; Therefore, the load balancing capability of the cloud workflow under the multi-tenant architecture is improved, so that the flow serviceprovider can provide services for more tenants on the basis of meeting the analysis execution performance requirements of different tenants on request throughput and different flow definitions.

The present application relates to a service data control method, device, computer equipment and storage medium. The method includes: acquiring test service data and preset curve types. Fit the test service data according to the preset curve type to obtain the throughput delay curve corresponding to the test service data. Determining the delay threshold corresponding to the test service data, and solving the curve function based on the delay threshold, to obtain the first throughput per unit time and the second throughput per unit time. Receive the target business data to be processed in sequence. Perform corresponding service processing according to the received target service data, and control the amount of target service data for service processing per unit time to be kept between the first throughput and the second throughput. By adopting the method, the system performance of the service data processing system can be stabilized.

The invention discloses a service balancing method based on base station throughput capability in a 5G network, comprising the following steps: step 1, calculating the real-time throughput of the base station; step 2, calculating the average real-time throughput of the base station according to the real-time throughput of the base station; Step 3, calculate the load balancing coefficient I of the network JF ; Step 4, every time T, recalculate the load balancing coefficient I of the network JF , according to the coefficient and the average real-time throughput of the network, the non-real-time services in the base station are preferentially adjusted.

The invention belongs to the technical field of mobile communication, and specifically relates to a method for regrouping workstations in a 5G network, including using the PS-Poll frame start transmission time of all workstations as an inspection indicator for hidden nodes to detect hidden node relationships in the network; After the access point obtains the PS-Poll frame transmission time of the workstation, it determines the hidden node relationship between the workstations and builds a hidden relationship matrix; the access point quantifies the expected service requirements reported by the workstations into different service levels, and builds the expected Business demand table; integrate the expected business demand table into the hidden node relationship matrix to form a hidden business table with a comprehensive understanding of the potential impact of hidden nodes; regroup workstations according to the detected hidden business table during each TBTT; The throughput of the STA regrouping method of the present invention is 40% higher than that of the existing random grouping method.

The invention provides an associated log playback method and device. The associated log playback method comprises the following steps: performing application classifying and business relating on application logs; and storing each application log performed the application classifying in sequence according to business associated relationships, wherein the business associated relationships comprise the follows: the beginning of the latter business depends on the completion of the former business; the application logs stored in sequence are divided into plurality groups of logs according to the number of press machines and the number of thread groups of each press machine, and each group of divided logs is transmitted to the corresponding press machine; each thread in the thread group of the press machine is controlled to read a bucket of logs in the corresponding log group successively and perform log sending at a preset sending speed, so as to meet the test throughput requirements; and the number of the press machines, the number of log groups and the number of thread groups are increased to carry out linear extension on the throughput when the sending speed is limited. By means of the invention, the associated logs can be replayed quickly and with high throughput.

The present invention provides a method and system for distributed calculation of user indicators in finance. The method includes: dividing the calculation of user indicators into three types of indicator calculations: offline calculation, business-triggered calculation, and data source change-triggered calculation; defining user indicators , including user indicator attributes and indicator operators. Indicator operators include defining the data source used in indicator calculation, indicator calculation logic, and indicator calculation type. There must be at least one indicator calculation type in the indicator operator; define triggers for three indicator calculation types conditions; according to the index calculation type to which the user index belongs, when the trigger condition occurs, the index calculation type corresponding to the trigger condition is used for index calculation; the calculated user index is stored, and the index calculation efficiency of the present invention is high, and can solve the problem of repeated index calculation and new Add / modify index calculation logic difficulties, serious index missing and other problems, and can effectively ensure the execution of high-quality computing and fast operators, and improve system stability and real-time throughput.

The invention discloses a smart grid resource management method and system based on time delay and throughput. The method includes: receiving and analyzing information on power services to obtain service requirements; Slice classification of services; calculate the downlink transmission rate corresponding to each slice, and calculate and obtain the real-time buffer queue length corresponding to each slice according to the downlink transmission rate; calculate the maximum downlink transmission rate of the eMBB slice The binary variable allocation scheme corresponding to the value and the optimal transmission power between the electric equipment and the gNodeB corresponding to each slice; according to the above results, the power is set for the electric equipment; the method and the system confirm the delay and throughput through analysis Restricted conditions, rationally allocate resource blocks of the smart grid, and maximize the throughput on the basis of ensuring the delay requirements, so as to realize the optimization of smart grid resource management.

The invention discloses a multi-priority scheduling method for industrial field data based on 5G slices, including S1: 5G base stations obtain resource request data and channel state data of industrial field users; S2: the scheduler of the 5G base station in the t-th scheduling period, Determine whether the task cache queue is empty; if it is empty, wait for the next scheduling cycle; if it is not empty, enter S3; S3: The scheduler of the 5G base station allocates the corresponding resources to different slices according to the resource request data of the user; S4: Perform resource scheduling for users in the same slice. The present invention completes resource allocation for the three major slices of 5G, ensures flexible scheduling and isolation of resources between slices, improves system throughput, and at the same time improves the fairness of scheduling methods for different business service quality requirements.

The invention discloses a business scheduling method based on a genetic algorithm to optimize the transmission rate weight, which mainly solves the problem that the existing method is difficult to comprehensively measure throughput and user fairness performance, and has poor adaptability to different systems. Including: 1) Obtain the data transmission rate set of the system; 2) Determine the fitness function of the average throughput of the system and the fairness among users; 3) Establish a genetic algorithm search model according to the fitness function; 4) Combining with the search model, simulate and determine each The optimal weight of the data transmission rate; 5) Obtain the user's data transmission rate, user parameters and business parameters; 6) Calculate the user scheduling priority; 7) Schedule resource blocks according to the scheduling priority, and repeat the execution for unallocated resource blocks The scheduling step completes the scheduling. The invention adopts the genetic algorithm to search for the optimal weight of each data transmission rate of the system, can flexibly change the fitness function, and effectively improves the adaptability of the business scheduling method to different systems.

The present invention discloses a service message forwarding method and device, wherein the method includes: selecting a link for service message forwarding this time according to the parameter information of each link in a virtual link binding group, wherein the virtual link binding group A link binding group includes one or more links of different link types; service packets are forwarded through the selected link. The present invention solves the problems of too long service forwarding delay and decreased service throughput caused by forwarding service messages according to links in turn in the prior art, achieves the technical effect of effective utilization of link resources, and improves The technical effects of business throughput, expanding link bandwidth, and increasing networking flexibility.