210 results about "Service-level agreement" patented technology

The invention discloses a computing power scheduling method and system based on dynamic load prediction and resource priority ranking. The computing power scheduling method comprises the following steps: collecting historical load data, task submission data and resource state data of each node in a computing power cluster; on the basis of the preprocessed multi-dimensional load feature data set, constructing an improved hybrid prediction model, optimizing model parameters through training, and predicting the load change trend of each computing power node in a future preset time period by using the trained model to obtain a node load prediction result; extracting a service level protocol parameter, a resource demand type and historical execution efficiency data of a to-be-scheduled task, and establishing a multi-dimensional resource priority evaluation index system; according to the computing power scheduling method, the problems of low resource utilization rate and high task response delay caused by low load prediction precision and mismatching of resource allocation and task priority in a traditional computing power scheduling method are solved, and the overall operation efficiency and service quality of a computing power cluster are improved.

The invention relates to the technical field of electric power communication, in particular to an electric power communication resource topological optimization method and system based on a graph database. The specific implementation process comprises the following steps: constructing an electric power communication resource topological graph comprising communication exchange nodes, data transmission links and attribute information through a graph database; analyzing and quantifying the received business type into a routing constraint condition according to a business service level protocol; performing dynamic weighting calculation on each candidate communication path meeting the routing constraint condition, evaluating comprehensive performance indexes and outputting an optimal communication path; and converting the optimal communication path into a topology configuration instruction capable of guiding network equipment to carry out data deployment, and carrying out dynamic optimization on the power communication resource topology. According to the method, the service quality requirements of different services are quantified into dynamically adjusted routing constraints and weights, so that reasonable allocation of power communication resources and effective optimization of network topology are realized, and the working efficiency of a power communication resource network is improved.

The invention provides a resource urgency degree evaluation and dynamic scheduling method and system based on behavior characteristics, and the method comprises the steps: collecting equipment protocol characteristic data, equipment resource state data and user historical behavior data, and carrying out the preprocessing, and obtaining a standardized data set; by fusing a convolutional neural network, a bidirectional gating circulation unit and a network model of an attention mechanism, extracting spatial-temporal characteristics and a long and short term dependency relationship of the standardized data set, and generating depth behavior characteristics; dynamically calculating a task urgency degree weight from three dimensions of business priority, an equipment performance baseline and environment health degree by adopting a fuzzy analytic hierarchy process based on the depth behavior characteristics, and generating a comprehensive urgency degree score; according to the comprehensive urgency score, generating a dynamic scheduling strategy and executing resource allocation; and monitoring the resource utilization rate and the service level agreement compliance rate, and feeding back for optimizing a subsequent scheduling strategy.

Techniques, methods, and solutions for supporting accurate and resilient positioning, navigation, and timing (PNT) services. In some examples, a cellular PNT (Cell-PNT) network may provide end-to-end monitoring and control of PNT capabilities. For example, a server may coordinate resource allocation between devices. The server may receive a service request and requested performance indicators according to a service level agreement. The server may additionally receive determined performance indicators associated with the service. Based on differences between determined performance indicators and requested performance indicators, the server may determine how to adjust resources to satisfy the requested performance indicators. Indication of the adjusted resources may be transmitted to a device associated with the determined performance indicators. The device may adjust data indications and signals based on the adjusted resources.

The invention relates to the technical field of load balancing optimization, in particular to an intelligent computing center load balancing and energy-saving optimization method and system, and the method comprises the following steps: obtaining task execution duration and resource state data, calculating a multi-dimensional index, carrying out the weighted analysis, dynamically outputting a regulation and control scheme, and achieving the load balancing and energy-saving optimization of an intelligent computing center. According to the method, response time comparison based on a service level protocol is introduced, a task execution duration distribution structure is clearly recognized, the abnormal task recognition precision is improved, and the judgment accuracy of an overload trend is enhanced by comparing resource utilization rate changes of adjacent periods in the aspect of node state monitoring; in the task allocation process, the available memory of the node and the resource adaptation degree are combined, real performance load performance is obtained through energy efficiency and throughput data cross analysis, task allocation better conforms to the actual energy consumption level, the resource waste risk is effectively reduced, the response flexibility of a scheduling strategy to system load changes is enhanced, and the task allocation efficiency is improved. And the task execution efficiency and the energy efficiency distribution accuracy are improved.

The invention discloses a data center uninterruptible power supply intelligent control method and device based on reinforcement learning, and relates to the technical field of energy management. The method comprises the steps that a health perception state space is constructed, and the state space comprises operation state features and standby battery multi-dimensional health vectors generated through training on a plurality of distributed uninterruptible power supply stations through a federated learning method; defining a multi-target reward function for comprehensively quantifying the economic cost based on the time-of-use electricity price, the standby duration satisfaction degree of the service level agreement and the battery life loss cost; through a reinforcement learning algorithm with a security constraint layer, based on the state space and the reward function, training an intelligent agent to learn an optimal power supply control strategy, the security constraint layer forcibly executing a preset operation security boundary in training and execution; and according to the learned optimal strategy, a real-time control instruction for the uninterruptible power supply is generated and executed.

This disclosure describes techniques for routing application traffic on a network path based on an application-specific Service Level Agreement (SLA) threshold that may be dynamically and / or automatically determined. An edge device of an SD-WAN may send probes over the SD-WAN over multiple paths and may determine network telemetry data representing the network performance associated with the multiple paths. The edge device may then send the network telemetry data to a control plane. The control plane may also receive quality of experience (QoE) data indicating the quality of an application experience for a user. Based on the network telemetry data and the QoE data, an application-specific SLA threshold may be determined for the application. The edge device may then route traffic of the application through a path of the multiple paths based at least in part on the path satisfying the SLA threshold.

This disclosure describes techniques for improving routing policy awareness in a network. The method includes detecting, by a controller, an application initiated for use at an edge node of a network. Then, generating, by an analytics engine coupled to the controller, analytical data of traffic flow at the edge node of the network wherein the traffic flow is in accordance with a routing policy for routing traffic associated with the application. Further, routing of the traffic through a path from one or more paths configured at the edge node that is in accordance with at least a Service Level Agreement (SLA) for traffic flow. Also, in response to an SLA violation during routing of the traffic, causing an action, by the controller, of routing traffic flow through another path that is in accordance with at least the SLA for traffic flow based on analytical data received of the traffic flow.

The invention discloses a dynamic resource allocation method and system based on 5G slices, and relates to the field of communication, and the method comprises the steps: collecting multi-domain network operation parameters and service portrait information; analyzing an access network resource utilization rate, a transmission link bandwidth, a core network queue length and slice quota information, and outputting a default probability of a slice service level protocol; calculating an elastic interval and a cross-slice borrowing matrix of slice resources in the digital twin model based on the default probability; based on the elastic interval and the cross-slice borrowing matrix, solving is carried out through a security reinforcement learning algorithm, and a target resource allocation vector and an action strategy are generated; and issuing the target resource allocation vector and the action strategy to an execution layer, controlling an access network scheduling weight, a transmission queue threshold and a core network session anchor point, and realizing slice resource dynamic allocation. According to the invention, the resource utilization rate is effectively improved, the cross-domain mismatching risk is reduced, and the service continuity and stability in a complex service scene are enhanced.

The invention discloses a load-aware large model reasoning service routing method and device, relates to the technical field of computers, comprehensively captures the real load condition of a reasoning execution unit through multi-dimensional operation state data, realizes the accurate quantification of the load state in combination with a multi-index weighted evaluation model, and then adapts to a service demand dynamic distribution request to realize the routing of the load-aware large model reasoning service. Therefore, the reasoning request can efficiently match the optimal resource, the imbalance problem that part of nodes are overloaded and queued and part of nodes are idle is avoided, and the effects of adapting to the characteristics of high computing power demand, large response time fluctuation and the like of the large model reasoning service are achieved. Therefore, on the premise of guaranteeing different user service level agreement (SLA) requirements, the cluster resource utilization rate and the service response efficiency can be remarkably improved, and meanwhile, the stability and the scene adaptation capability of the large model reasoning service are enhanced.

The invention relates to the technical field of internet risk collaborative management, and discloses an internet data risk early warning and collaborative management system and method.The internet data risk early warning and collaborative management method comprises the steps that a service grid enhanced federal learning unified framework is constructed, and a tetrad structure is formed; a seamless integrated architecture of federal learning and a service grid is realized; developing a federal contribution evaluation mechanism based on a service level agreement; a self-adaptive service discovery and dynamic combination technology is realized, and an optimal service combination scheme is generated; constructing a federated aggregation algorithm of data quality perception, and introducing a quality weight factor optimization model for training; designing a multi-level cache strategy; according to the invention, through the fusion of federal learning and the micro-service architecture, the purpose of keeping efficient collaboration while protecting data privacy is realized, the original data of each participant is always kept locally, only necessary model parameters are shared, and the privacy leakage risk is reduced.

In one embodiment, a device obtains information regarding temporary routing patches applied to a network. Each temporary routing patch implements a routing change in the network for a specified amount of time to avoid or mitigate against a service level agreement violation. The device evaluates, using the information regarding the temporary routing patches applied to the network, a plurality of replay scenarios for the network. The device determines, based on the plurality of replay scenarios, a long-term configuration change for the network. The device provides an indication of the long-term configuration change for display.

The invention relates to a resource allocation method and device, equipment, a storage medium and a computer program product. The method comprises: receiving a resource allocation request of a target service; when the isolation level is a preset isolation level, acquiring network resource state information of the current network, and inputting the priority, the resource demand and the network resource state information corresponding to each network slice into a resource allocation model to obtain a target resource allocation result; the resource allocation model is used for executing a resource allocation strategy taking minimization of resource waste as an optimization target and satisfying service level protocol parameters corresponding to a plurality of network slices as constraint conditions; and carrying out resource allocation on each network slice based on the target resource allocation result. According to the content, the resource waste of each network slice can be minimized, so that the resource utilization rate is improved.

A system and methods are provided for improving power efficiency of a data center, including: acquiring training data including power caps, utilization rates, and a measure of Service Level Agreement (SLA) compliance of one or more computer servers of the data center; creating a model for determining power caps according to measured utilization rates of the one or more computer servers, wherein the determined power caps, when applied to the one or more computer servers, reduce power consumption and meet the measure of SLA compliance; and applying the model, according to subsequent data received during a second operating period, to determine a power cap to apply to the one or more computer servers, wherein the subsequent data includes a subsequent utilization rate of the one or more computer servers.

The invention discloses a server computing power resource dynamic scheduling method based on multi-objective optimization, and relates to the technical field of server scheduling, and the method comprises the steps: obtaining real-time operation state data of a server cluster and attribute information of a to-be-scheduled computing task in response to a dynamic scheduling trigger event; based on the real-time operation state data and the attribute information, constructing a multi-objective optimization model; the multi-objective optimization model is solved, and a Pareto optimal solution set representing different resource allocation schemes is obtained; and selecting an optimal scheduling scheme from the Pareto optimal solution set based on a preset operation strategy. The method has the advantages that the dynamic multi-objective optimization model fusing the performance, the energy efficiency and the resource utilization rate is constructed, and online solving is performed by adopting the decomposition-based evolutionary algorithm, so that comprehensive optimization and optimal balance of the computing power resource utilization rate, the task processing efficiency and the overall energy efficiency of the system are realized on the premise of ensuring the service level agreement.

The disclosure relates to a 5th generation (5G) or pre-5G communication system for supporting a data transmission rate higher than that of 4th generation (4G) communication systems, such as long term evolution (LTE). A method of performed by a radio access network intelligence controller (RIC) for guaranteeing a service level agreement in a wireless communication system is provided. The method includes transmitting, to an open radio access network distributed unit (O-DU), a subscription request message including configuration information for reporting whether the service level agreement is satisfied for each network slice or network slice group, receiving, from the O-DU through an E2 interface, an instruction message including information about whether the service level agreement is satisfied, and transmitting, to the O-DU, a control message including scheduling information set based on the instruction message.

The invention provides an intelligent scheduling management system based on a carbon footprint model, and the system achieves the dynamic coupling of calculation task scheduling and the real-time carbon intensity of a power grid through the cooperation of a carbon footprint modeling engine, a dynamic scheduling controller, a resource state sensing layer and a feedback execution module. The problem that a traditional data center resource scheduling system is lack of carbon sensing capacity is effectively solved. By establishing the dynamic association between the power source and the geographic position and fusing the multi-dimensional equipment energy consumption characteristics and the task demand labels, the scheduling decision automatically responds to the fluctuation of the regional renewable energy sources, and the overall carbon footprint is remarkably reduced on the premise of guaranteeing the calculation performance and the service level agreement. Meanwhile, the improved multi-objective optimization algorithm can efficiently search for an optimal balance point in a solution space formed by carbon emission, default rate and economic cost, the resource energy efficiency utilization rate is further improved through an elastic expansion mechanism based on real-time state perception, and finally unification of data center operation environment benefits and economic benefits is achieved.

Methods, apparatus, systems, and articles of manufacture are disclosed to cache media based on service level agreement (SLA) type. An example apparatus includes caching function execution circuitry to evaluate a caching function with (A) a type of an SLA corresponding to a profile associated with a device requesting the media and (B) metadata corresponding to an entry of the cache corresponding to the media, the profile corresponding to an identification (ID). The example apparatus also includes caching policy circuitry to, in response to the caching function indicating that the metadata satisfies the SLA, cause transmission of the entry of the cache to the device that requested the media.

The invention discloses a network operation and maintenance automatic management method based on deep learning, and the method comprises the following steps: collecting operation and maintenance data, and carrying out the preprocessing of the operation and maintenance data, and obtaining a standardized data set; adopting a Mama selective state space network to generate an operation and maintenance state vector; inputting the operation and maintenance state vector into the improved CORAL model to generate an anomaly detection result; generating a root cause analysis result by adopting a causal inference method; generating candidate operation and maintenance actions by adopting a constraint modeling method in combination with a service level agreement rule; obtaining an optimal operation and maintenance action by adopting an action evaluation method and a risk analysis method, issuing and executing the optimal operation and maintenance action, and collecting execution feedback data; and generating an operation and maintenance audit report based on the anomaly detection result, the root cause analysis result and the execution feedback data. Intelligent, automatic and auditing management of the whole operation and maintenance process is realized, and the method can be widely applied to intelligent operation and maintenance scenes of a large-scale complex network.

Multi-Access Point (AP) coordinated Service Periods (SPs) may be provided. Flow-set data is determined comprising data for one or more Quality of Service (QoS) flows-sets of one or more clients. The flow-set data is sent to one or more neighboring co-channel APs. Neighbor flow-set data is received comprising data for one or more neighbor QoS flow-sets from at least one of the one or more neighboring co-channel APs. Service Level Agreements (SLAs) are established for the one or more QoS flow-sets and the one or more neighbor QoS flow-sets based on the flow-set data and the neighbor flow-set data. One or more SPs are scheduled for at least one of the one or more QoS flow-sets and the one or more neighbor QoS flow-sets based on the SLAs.

Disclosed is a method (500) for adjusting Service Level Agreement (SLA) parameters in a communication environment (100). The method includes collecting 5 data associated with workload, resource utilization, and performance metrics. The method further includes setting one or more SLA parameters for one or more services included a SLA. Further, the method includes predicting a future workload pattern, resource requirements, and future SLA breaches by analyzing the collected data. Furthermore, the method includes evaluating a performance of the one or more 10 services against SLA target and the future workload pattern. Furthermore, the method includes adjusting the SLA by adjusting the one or more SLA parameters in real-time to maintain a consistent service level in the communication environment.

The invention relates to the technical field of electric power inspection management, and particularly discloses an electric power system inspection method based on machine learning, which comprises the following steps that: an edge end firstly reads self-inspection data of a sensing device in a historical inspection period and judges whether the self-inspection data are available or unavailable, so that only an original stream is extracted from qualified data and a window observation set of the same equipment in the same time period is constructed; sending the window observation set and the availability state into a lightweight model to generate an initial defect vector; then, according to an uplink strategy, only generating an associated data packet for suspicious or inaccurate samples, and uploading the associated data packet; the cloud carries out regression fusion on uplink samples to obtain defect probability and execute gating, risk assessment is carried out according to event tags to output risk scores, finally, the risk scores are used for limiting a self-adaptive inspection cycle and automatically configuring a next round of inspection and acquisition modality, high-risk encryption, low-risk frequency reduction, reduction of invalid mileage and repeated order sending are realized, and the accuracy of inspection is improved. And the service level agreement achievement and overall operation and maintenance efficiency is improved.

The invention relates to the technical field of cloud and mist collaboration, in particular to a cloud and mist collaboration application dynamic deployment method and system based on load balancing, and the method comprises the steps: collecting dynamic resource data of a fog node in real time through a lightweight resource monitoring agent deployed at the fog node, based on the historical access log of the terminal equipment and the service level agreement of the manufacturing application, calculating the delay sensitivity and access popularity of each application; generating a priority queue of application deployment according to a preset priority formula, wherein the priority formula is determined based on the delay sensitivity and the access popularity; allocating the applications to the fog nodes meeting the resource capacity according to the descending order of the priorities until the fog node resources are fully occupied; after fog node resources are fully occupied, if it is determined that undeployed applications exist, a Pareto optimal solution set meeting end-to-end delay constraints is generated, and a scheme with the minimum resource utilization rate variance is screened to dynamically deploy the applications; according to the invention, global optimization of cloud and fog resources can be realized.

Techniques, methods, and solutions for supporting accurate and resilient positioning, navigation, and timing (PNT) services. In some examples, a cellular PNT (Cell-PNT) network may provide end-to-end monitoring and control of PNT capabilities. For example, a server may coordinate resource allocation between devices. The server may receive a service request and requested performance indicators according to a service level agreement. The server may additionally receive determined performance indicators associated with the service. Based on differences between determined performance indicators and requested performance indicators, the server may determine how to adjust resources to satisfy the requested performance indicators. Indication of the adjusted resources may be transmitted to a device associated with the determined performance indicators. The device may adjust data indications and signals based on the adjusted resources.

Various systems and methods for reactive intent-driven end-to-end (E2E) orchestration are described herein. An orchestrator system, includes a processor; and memory to store instructions, which when executed by the processor, cause the system to: receive, at the orchestrator system, an intent-based service level agreement (SLA) for execution of a series of tasks on a plurality of compute nodes; calculate, based on the intent-based SLA, intermediate latency thresholds corresponding to each task of the series of tasks; calculate slack estimates based on the latency thresholds and real-time telemetry of the plurality of compute nodes or real-time telemetry of connections between the plurality of compute nodes; monitor execution of the series of tasks on the plurality of compute nodes; and perform a corrective action in response to determining that the execution of the series of tasks is predicted to exceed one of the intermediate latency thresholds.

Controlling carbon emissions by workload at the storage level is provided. A storage system that can run a workload within a remainder of a specified time period without exceeding a maximum carbon dioxide (CO2) equivalent emission threshold level defined for the workload in a sustainability service level agreement (SLA) is identified based on analyzing retrieved CO2 equivalent emission data in a time-series format from a plurality of storage controllers corresponding to a plurality of storage systems using a set of machine learning models. The workload is migrated to the storage system prior to the specified time period being met to decrease CO2 equivalent emissions of the workload in accordance with the sustainability SLA.

A method may include, obtaining, by one or more processors, transaction logs of a database server captured in a predetermined time period, identifying, by the one or more processors, a rate of change of data within the transaction logs, classifying, by the one or more processors, portions of the time period according to the rate of change of data within the portions of the time period, and setting, by the one or more processors, a service level agreement (SLA) for each portion of the time period according to the classification of each portion.

An apparatus for evaluating the quality of digital human content included in a source input are disclosed. The apparatus includes a first entity determines a provisioning plan based on available resources matching a service level agreement (SLA)-based user's provisioning request. The apparatus further includes a second entity provision resources of the QKDN based on the provisioning plan, and re-provision the provisioned resources in response to receiving a re-provisioning request or a re-optimization request for the provisioned resources from other entities.

The invention provides a dynamic data-backup scheduling system. The system includes a memory storing one or more processor-executable routines and a processor communicatively coupled to the memory. The processor is configured to execute one or more processor-executable routines to determine a dynamic backup schedule for one or more computing machines based upon data change patterns of the respective computing machines. The processor is further configured to access user inputs corresponding to data backup for the one or more computing machines. The user inputs comprise at least one of a tolerable data loss threshold, critical data type, file sensitivity data, and exceptions data time period information. The processor is further configured to create one or more service level agreement (SLA) configurations for data backup based on the user inputs. The processor is further configured to access historical data corresponding to one or more data backup cycles of the one or more computing machines. The historical data comprises one or more of time series changes, the addition and / or change of data information for pre-determined data sets, and corresponding file metadata of the data sets. The processor is further configured to implement one of the active and passive modes of data backup using the user inputs and the historical data. The processor is further configured to trigger a backup job for one or more computing machines based on a comparison between the SLA configurations and dynamic information received about data addition and / or data change of the computing machines in the active mode. The processor is further configured to trigger a backup job for one or more computing machines based on a comparison between the SLA configurations and historical time series change and / or addition of data of the computing machines in the passive mode.

Examples described herein relate to a configurable switch with dynamically configurable device connections to a processor socket, where the device connections are configured to meet service level agreement (SLA) parameters of a first service executing on the processor socket. For a second service that is to execute on the processor socket and the second service is higher priority than the first service, device connections of the switch to the processor socket are dynamically reconfigured to meet SLA parameters of the second service.