40 results about "Elastic scheduling" patented technology

The invention discloses a DCOS-based cloud management platform. The platform comprises a resource management module, a monitoring management module, an elastic capacity expansion and reduction scheduling module, a unified log center, a continuous integration platform and an operation and maintenance management module, and the platform is characterized in that the resource management module comprises arrangement, network management, host management and storage management; the monitoring management module comprises authority management, monitoring and logging; the elastic capacity expansion andreduction scheduling module comprises elastic expansion and reduction and load balance; the unified log center comprises micro-service operation and maintenance management; the continuous integrationplatform comprises deployment, a template and mirror image management; and the operation and maintenance management module comprises a module center, configuration management and operation maintenancemanagement. According to the invention, the platform achieves the dynamic IT resource allocation, technical platform integrated management, rapid elastic deployment management and stronger IT management and control, achieves the purpose of managing the data center resources by using the DCOS, achieves the elastic scheduling of the resources, reduces the difficulty of system operation and maintenance, enables the IT resources to be allocated according to requirements, and saves the resources.

The invention discloses a method and a system for elastically scheduling resources. The method comprises the following steps: acquiring target data, and performing data analysis on the target data toobtain a functional index for representing a service running condition; obtaining a service demand according to the functional index; and acquiring an alarm event and an elastic scheduling strategy, and performing elastic expansion and reduction adjustment conforming to the service demand on container resources in a data center resource pool according to the alarm event and the elastic schedulingstrategy.

The invention provides a timing method for traffic signal lamps, and belongs to the field of transportation .. / .. / .. / Program Files / Youdao / Dict / 7.5.2.0 / resultui / dict / keyword=facilities. The method comprises the following steps: conducting single-crossroad signal lamps timing for the crossroad in case of only one crossroad; conducting single-crossroad signal lamps timing for each crossroad if the distance between two adjacent crossroads is greater than 800 meters; conducting signal lamps timing for the two adjacent crossroads if the distance between two adjacent crossroads is less than or equalto 800 meters; and conducting signal lamps timing for the four crossroads if a crossroad is adjacent to four crossroads. The timing method for traffic signal lamps applies a real-time elastic scheduling strategy to the crossroad signal timing for different lane saturations, and constructs the timing model for a variety of crossroads, solves a corresponding optimal analytical solution, and designsan optimal green lamp timing for each phase, so as to fundamentally relieve traffic pressure.

The invention provides a container cross-host online migration method, a storage medium and terminal equipment. The method comprises the following steps of carrying out resource monitoring on each physical node of the cloud platform; setting a migration threshold as a judgment condition for triggering container migration; selecting a container with the minimum migration cost as a container to be migrated; determining a physical node of which the comprehensive load value is smaller than a set value, calculating an average value of the comprehensive load values of the physical nodes; selecting aphysical node with a comprehensive load value closest to the average value as a target node; and executing migration until the comprehensive load value of a certain node is smaller than a migration threshold value, so that load balance of each physical node in the cluster can be maintained, services in the container are not interrupted while the container migrates, and thus elastic scheduling ofDocker cloud platform resources is realized.

The invention discloses a gas-electric coupling comprehensive energy system planning method considering an elastic scheduling strategy. The method comprises the following steps: inputting the structure and equipment parameters of a gas-electric coupling comprehensive energy system according to the structure of the gas-electric coupling comprehensive energy system; establishing gas-electric coupling comprehensive energy system equipment operation constraints of the cooling typical day and the heating typical day; establishing a multi-stage optimization scheduling model oriented to elastic operation; establishing an optimization model based on a genetic algorithm; and carrying out interactive solution on the optimal configuration layer multi-stage optimal scheduling model and the operation optimization layer optimization model and outputting a solution result, wherein the solution result comprises an equipment configuration scheme and an optimal scheduling plan. According to the invention, the complementary coordination effect of the electric drive equipment and the gas drive equipment and the response capability of the energy storage device are fully considered; and a gas-electric coupling comprehensive energy system double-layer planning and design model considering an elastic scheduling strategy is established, the upper layer is evolved and optimized by adopting a genetic algorithm, and a related mathematical solver is called for solving the lower layer to obtain a system planning and design scheme and an optimal scheduling planGas-gas scheduling strategy considering elastic scheduling strategy Electrical coupling comprehensive energy system planning method: gas-gas coupling comprehensive energy system planning is carried out according to gas-gas Electrical coupling integrated energy system structure and input gas- The structure and equipment parameters of the comprehensive energy system are electrically coupled; establishment of gas-heating typical day and cooling typical day Equipment operation constraint of the integrated energy system is electrically coupled; establishing a multi-stage optimization scheduling model oriented to elastic operation; establishing an optimization model based on a genetic algorithm; and carrying out interactive solution on the multi-stage optimization scheduling model of the optimization configuration layer and the optimization model of the operation optimization layer, and outputting a solution result which comprises an equipment configuration scheme and an optimal scheduling plan. Complementary coordination effect of electric driving equipment and gas driving equipment and response capability of an energy storage device are fully considered, and a gas-gas scheduling strategy considering an elastic scheduling strategy is established; and a double-layer planning design model of the integrated energy system is electrically coupled, an upper layer adopts a genetic algorithm to evolve and optimize, a lower layer calls a related mathematical solver to solve, and a system planning design scheme and an optimal scheduling plan are obtained.

The invention discloses a flexibly-scheduled network and service monitoring system based on a cloud platform. The system comprises a cloud-side module and an enterprise-side modul the information communication is performed by establishing a VPN (virtual private network) between a cloud monitoring platform access gateway of the cloud-side module and an enterprise monitoring system outlet gateway of the enterprise-side module; the cloud-side module comprises a user management module, an information issuing platform, an alarm management module, a configuration management module, a report management module, a component pool module, an acquisition service module and a database module; a plurality of server nodes are arranged at the enterprise-side module, each server node comprises an Agent module, an operation monitoring module, a local scheduler module, a component manager module and a local component pool module. The invention puts forward a network and service monitoring system which centralizes the monitoring alarms, centralizes the configuration issuing and performs the full-management based on the cloud platform, and has the following features: (1) cloud platform performs the centralized control, and the back-end professional terminal service is provided; (2) the remote scheduling issuing is performed and the deployment is as required; (3) function is flexible in expansion, and the system operation state can be mastered anytime anywhere through the short message, SMS and APP.

The invention relates to a system and method for intelligently and flexibly scheduling data center resources. The system comprises an operation and maintenance monitoring module, a data analysis module, a flexible scheduling module and a strategy maintenance module, wherein the operation and maintenance monitoring module is used for collecting performance and log data and generating a corresponding alarm event according to the category and operation condition of an application program; the data analysis module is used for obtaining the alarm event, judging whether to carry out flexible scaling according to a preset flexible scheduling strategy, and if so, sending a first flexible scaling instruction to the flexible scheduling module; and the flexible scheduling module is used for responding to the first flexible scaling instruction to judge whether a data center resource pool satisfies flexible scaling requirements, if so, sending a second flexible scaling instruction to the data center resource pool and recording a flexible scaling event of the data center resource pool. Compared with the prior art, the system provided by the invention can carry out flexible expansion and contraction according to service demands, thereby improving the stability of a data center.

The application discloses a flexible scheduling method and device for database comprising following steps: obtaining the performance parameter values of a database server and database instances that participate in the flexibility scheduling; selecting a database server whose performance parameter value exceeds a preset capacity upper limit; selecting a database instance from the database server and migrating it to a database server capable of accepting; determining whether or not a predetermined scheduling termination condition is satisfied, and if not, turning to and executing the step of selecting the database server whose selection performance parameter value exceeds the preset capacity upper limit. The application provides a flexible scheduling device for database at the same time. The use of the flexible scheduling method for database provided in this application avoids the jitter caused by the real-time change of the performance of the database instance, on the other hand can reduce the number of migrating database instances, reduce the migration cost, and use the heuristic method to avoid the complicated calculation process, effectively improve the efficiency of the implementation.

The invention discloses a transformer area flexible scheduling method and device based on flexible load dynamic characteristics. The method comprises the following steps: establishing a multi-objective optimization model; solving the multi-objective optimization model by adopting a multi-objective particle swarm optimization algorithm to obtain an optimal solution set; and carrying out peak load shifting scheduling on the load according to the schedulable resources of electric vehicles in the optimal solution set and the maximum reducible quantity of air conditioners. The transformer area flexible scheduling method and device based on the flexible load dynamic characteristics have the advantages that peak clipping and valley filling are achieved, the power utilization satisfaction degree of users is improved, and the charging resource utilization rate is maximized.

The invention discloses a virtual resource flexible scheduling implementation method combining application performance indicator monitoring data. The method includes the following steps: 1), scheduling the configuration which needs elastic pool scheduling rules according to time and / or business system indicators; (2), conducting scheduling based on elastic pool scheduling rules; 2-1), declaring virtual machine queue to be started and stop queue; 2-2), inquiring the available elastic pool; 2-3), conducing circulation processing; 2-4), judging the scheduling rules, executing the time scheduling method of 2-4-1) when the rule is the time scheduling rule; executing the business system indicator scheduling method of 2-4-2) when the rule is the business system indicator scheduling rule; 2-5), starting all virtual machines of the queue to be started; 2-6), stopping all the virtual machines of the queue to be stopped; and 2-7), terminating the scheduling. The virtual resource flexible scheduling implementation method combining the application performance indicator monitoring data solves the problem that the hardware resource is wasted in the prior art, adopts the flexible scheduling rules, reduces the waste of the hardware resource by the business system, and effectively reduces the cost.

The embodiment of the invention provides an elastic scheduling method and system, and the method comprises the steps: collecting a time sequence index of a key feature of a service in a preset time period from a server, wherein the time sequence index of the key feature is an index value sequence of the key feature with a time index; respectively calculating the weight of each key feature and the quantile of each key feature according to the time sequence indexes of the collected key features; and according to the time sequence index newly generated by any key feature of the service, adopting the quantile of the key feature and the weight of the key feature to judge the time sequence index newly generated by the key feature, obtaining a final scheduling behavior decision, and carrying out scheduling according to the final scheduling behavior decision, wherein the scheduling behavior decision comprises one of the following decisions: index abnormity alarm and automatic capacity expansion and contraction. The time sequence indexes of the key features in the historical data are processed to judge whether to schedule or not, so that the inaccuracy of judging whether to schedule or not due to relatively strong subjective factors of manually determining the threshold value is avoided.

The invention relates to a hybrid resource dynamic scheduling method in a heterogeneous cloud computing environment, which comprises the following steps of: 1, creating a resource cluster according tobusiness requirements, and configuring available computing hosts of the same virtual machine type; Step 2, based on the created resource cluster, creating an available domain according to a service demand, and configuring an available cluster of the same virtual machine type; Step 3, based on the created available domain, creating a resource pool according to business requirements to set computing resources, storage resources and network resources, and adopting an elastic scheduling strategy by default for the newly created resource pool; And step 4, based on the created available domain, theuser applies to create the virtual machine on the console.

The invention discloses a GPU resource elastic scheduling method based on a heterogeneous application platform. The method comprises the following steps: a) obtaining GPU resource utilization rate information; b) setting a trigger threshold value and the number of times; c) screening and sequencing the capacity-reducing platform queues; d) screening and sequencing the expansion platform queues; with the steps of 1) selecting a to-be-reduced platform; 2) establishing a GPU node list; 3) processing a locking state node; 4) offline the to-be-migrated node; 5) adding to a resource queue; and 6) judging whether the volume reduction is finished or not. According to the GPU resource elastic scheduling method, flexible adjustment can be conducted according to the GPU load condition of the whole platform, so that maximized utilization of platform GPU resources is achieved, scheduling of the platform is mainly achieved through an existing scheduling assembly of an adaptive platform on the lower layer, dynamic resource monitoring, information collection and execution operation issuing are achieved through interface calling. And rapid and flexible deployment and implementation of cloud computing, big data, artificial intelligence and high-performance computing scene platforms can be satisfied.

Gas-for-use considering energy storage The elastic scheduling method for the electric coupling comprehensive energy system comprises the following steps: selecting a gas-to-gas system according to selected gas-to-gas; electrical coupling integrated energy system and input gas- The structure and parameters of the comprehensive energy system are electrically coupled; establishment of gas Electrically coupling the comprehensive energy system equipment operation constraint and the cold / power supply and demand balance constraint; calculating an energy storage standby capacity rolling calculation stage; optimal scheduling of the day-ahead economic scheduling stage is carried out; and carrying out optimization scheduling of a real-time operation stage. The invention is based on solving the problem of gas-to-gas The optimal scheduling problem of an electric coupling comprehensive energy system is solved, the complementary coordination effect of electric driving equipment and gas driving equipment is fully considered, the response capacity of an energy storage device under the emergency condition is brought into play, and energy storage standby gas-gas is established and considered. And electrically coupling the multi-stage elastic scheduling model of the integrated energy system to obtain a system day-ahead scheduling plan and a load recovery scheme when a fault occurs.

The invention discloses a containerized service scheduling framework and an elastic scheduling algorithm. The algorithm comprises the steps that a user request is subjected to semantic analysis and forwarding through an API gateway of a main node; the analyzed actual operation objects are created and scheduled, and all the objects are stored in a key value database; additional controller processes perform health monitoring on the objects in the cluster. According to the invention, a Docker container technology and a Kubentes container management technology are integrated, a power Internet of Things edge side service scheduling framework based on the container technology is provided, and under the power Internet of Things edge side service scheduling framework based on the container technology, in order to make up for the defects of a power Internet of Things edge side service scheduling algorithm, on the basis of an annotated ant colony algorithm, a power Internet of Things edge side service scheduling algorithm is provided. In order to optimize and improve initialization and updating of pheromones, setting of heuristic factors and the like, an elastic resource dynamic scheduling algorithm based on an ant colony algorithm is designed, load balancing of edge end task allocation is achieved, and limited edge resources are made to play the maximum effect.

The invention discloses an online second-hand vehicle source elastic scheduling method and system and a scheduling server. According to the embodiment of the invention, according to the SKU vehicle condition, the SKU display board price and the local SKU vehicle condition of the same type, the price of the local SKU display board of the same type, the ranking sequence of the SKUs is determined, the positions of the SKU units scheduled on a user client page are determined jointly through various data, the scheduling accuracy is greatly improved, the SKU units with high cost performance are fully displayed, and a second-hand vehicle source database is effectively utilized. Meanwhile, the price of the SKU display board is calculated; the SKU display board price can guide a merchant to provideSKU display board price debugging for the merchant, the method can effectively guide the merchant to adjust the SKU display board price to a proper price, and finally play a role in scheduling the SKU unit at the page position of the client, and the second-hand vehicle source database is further effectively utilized.

The invention discloses an automatic elastic task scheduling method, which realizes task scheduling by a Java timed task scheduling tool, and comprises the following steps of: before closing a service, storing or updating a scheduling task and related parameters thereof into a database; when the service is restarted, the stored scheduling task is taken out of the database, and if the scheduling task is triggered next time, the scheduling task is added into the task scheduler to be rescheduled after the rescheduling triggering time of the scheduling task is obtained according to the stored relevant parameters. According to the method, on the premise of ensuring the elastic scheduling, the service task can be closed without losing, and if the service is closed and the tracker is missed, thenext tracker time can also be calculated by restarting the service.

A method for scheduling resources in a network where the scheduling activity is split across two nodes in the network is disclosed, comprising: receiving, from a local scheduler in a first radio access network, access network information at a global scheduler; accessing information regarding a second radio access network allocating, at the global scheduler, resources for secondary allocation by the local scheduler; applying a hash function to map the allocated resources for secondary allocation to a set of hash values; and sending, from the global scheduler, the set of hash values to the localscheduler.

The invention discloses a cloud network resource elastic scheduling system and method based on an integral algorithm. The system comprises a virtual switch module, a data acquisition module, an integral calculation module and a resource scheduling module. The virtual switch module serves as a core module for bearing a VM network function and is used for providing network service for the VM according to a CPU cycle proportion; the data acquisition module is used for collecting the actual CPU clock cycle number and the cycle proportion of the virtual switch consumed by each VM in each working time slice from the virtual switch module, and transmitting the actual CPU clock cycle number and the cycle proportion to the integral calculation module; the integral calculation module is used for updating the integral value of each VM according to the input actual CPU clock cycle number and the cycle proportion, and transmitting the updated integral value to the resource scheduling module; and the resource scheduling module is used for dynamically limiting the number of virtual switch CPU cycles consumed by each VM in the next working time slice and the cycle proportion according to the input integral value. According to the method, resource utilization of a plurality of VM networks tends to be at a reasonable level.

The invention belongs to the technical field of cloud computing, and provides a method for flexibly scheduling virtual resources. The method includes S1, acquiring node state data of node sets in running states by the aid of a state acquisition module and storing the node state data in databases; S2, transmitting migration instructions to nodes by the aid of a resource scheduling module when scheduling strategies are triggered by the nodes in the node sets in the running states, and executing migration operation by the aid of a migration module; S3, enabling the resource scheduling module to judge whether the nodes are in idle states or not, closing the nodes if the nodes are in the idle states, adding the closed nodes to node sets which wait to be in running states, or preventing the nodes from being closed if the nodes are not in the idle states. The method has the advantages that the virtual resources in computer cluster service can be reasonably and flexibly scheduled by the aid of the method and accordingly can be prevented from being blindly configured, and energy consumption and the system overhead of cluster servers can be reduced.

The invention provides an evaluation method and device of elastic property, and relates to the technical field of the cloud computation. The method comprises the following steps: detecting the CPU usage rate of a detected server at the current moment; controlling a load generator to increase the load at the detected server under the situation that the detected CPU usage rage is greater than the preset threshold value, thereby triggering the detected server to start the elastic scheduling strategy; acquiring an elastic scheduling parameter in the process of starting the elastic scheduling strategy by the detected server, wherein the elastic scheduling parameter comprises the elastic scheduling time of the elastic scheduling strategy; and determining the elastic expansion capability of the detected server based on the elastic scheduling parameter. A technical problem that the testing way of the elastic property is poor in accuracy in the prior art is relieved.

The invention discloses a tensorflow-based multi-task elastic scheduling method and system, and the method comprises the following steps: carrying out the preprocessing of tasks entering a cluster through a task management system, and allowing a plurality of tasks in the cluster to be carried out at the same time; obtaining the number of all GPU resources in the cluster from a configuration file of the task management system, dividing a data set for the training task by utilizing the number of available GPUs, and dividing each part of data set to a specified GPU; elastically allocating video memory resources to the tasks on each GPU, and elastically expanding idle GPUs to improve the training speed; completing gradient descent on each part of data set to obtain the current gradient of eachpart; accumulating the gradients by using the communication among the clusters to obtain a current total gradient; broadcasting the total gradient to each GPU, and then performing next gradient descent. According to the method, multiple tasks can be elastically scheduled to enter the cluster, and distributed computing is efficiently completed by utilizing existing resources of the cluster.

The invention discloses a quota elastic scheduling method for an AI computing cluster, which is applied to a cloud platform. The method comprises the following steps: configuring a scanning interval, a capacity expansion threshold and a capacity expansion strategy according to the cloud platform; scanning the computing resources of the container running on the cloud platform, and scanning the computing resources of the cloud platform; performing maximum verification on a first computing resource of a first computing task which is running in the container and a second computing resource of a second computing task which is about to run in the container; judging whether the computing resources of the container meet maximum verification or not, and if not, executing a capacity expansion strategy step; detecting whether the computing resources of the cloud platform reach the capacity expansion threshold value or not: if yes, starting the capacity expansion strategy for the container. In this way, when the cloud platform has many idle computing resources, the computing resources can be flexibly allocated to enterprise users, the users in need can fully utilize the idle resources to carry out tasks, and the cluster resource utilization rate is also improved.

The invention provides an intelligent single image signal indicator device. The device is an indicator device which is combination of a single image and the tablet computer technology and is used for indicating traffic signals. A display module, a monitoring module and a communication module are arranged in a tablet computer system, the display module comprises single image display area, lamp signals and traffic signals with different colors are displayed through built-in software transmission signals, the display area can be enlarged since there is a single display image, more clear and visible lamp signals can be provided for pedestrians, the monitoring module feeds back road information in real time and can be used as an intersection monitoring system due to elastic scheduling in traffic rush hours for maintenance of social security. According to the device, the tablet computer technology is integrated with a conventional traffic signal indicator, a shoulder signal controller and the intersection monitoring system so that simplification, system cloudification, and intelligence of the traffic signal indicator can be realized.

The embodiment of the invention provides a cloud application scheduling method and device, electronic equipment and a storage medium, and is applied to a cloud application scheduling process, the method comprises the following steps: obtaining a cloud application provided for a user to use, and abstracting an operation unit required by the cloud application to obtain an instance; wherein the instances are obtained by uniformly abstracting operation units based on different application ecology, and the instances comprise resource occupancy; and according to the resource occupation amount of the instance and the use condition of the current instance, performing elastic scheduling on the instance under any application ecology. According to the method, the unified elastic scheduling capability of different application ecology cloud applications is realized by utilizing the bottom layer abstract capability of different application ecology, so that the development cost of an elastic scheduling function of a cloud application service manufacturer can be reduced, unified user experience can be provided for a cloud application manager, and the understanding cost is reduced.

The invention relates to the field of data processing application. The invention discloses a distributed resource elastic scheduling model and method, electronic equipment and a storage medium. The distributed resource elastic scheduling model and method realize cooperative work of task layer dynamic resource scheduling and system layer resource scheduling, give a response to the resource demand of the stream processing task in real time, allocate reasonable physical resources, achieve automatic elastic resource scheduling in the true sense, elastically increase and decrease physical resourcesoccupied by the distributed stream processing system, meet the resource requirement that stream processing tasks change at any time, guarantee the stable operation of the distributed stream processing system and the stable and efficient work of the stream processing system, and meanwhile, increase the resource utilization rate and reduce the cost.

A dynamic fault-tolerant elastic scheduling method comprises the following steps: estimating a task running time; determining a scheduling type of the task according to a flexible time situation of the task; according to the scheduling type of the task, selecting the virtual machine executing the task from the virtual machine list and scheduling to realize the reliability and resource optimizationbetween the clouds.

The invention is an elastic control optimizing method for bio-fermentation batch course. On the condition of ensuring the top-bottom working segments and load stable, firstly determining an elastic control interval for the cylinder batch to be stopped, where the optimum time of stopping cylinder batch is; then making on-line calculation of an elastic control function of the cylinder batch in this interval, where the function quantitatively gives the benefit-making potential of the cylinder batch relative to average cylinder batch; and finally according to the function, implementing the object of the maximum economic benefits and applying the policy of elastically stopping cylinder to determine the optimum cylinder batch-stopping time. It stops low-benefit and -potential cylinder batch in advance, thus reducing economic losses; prolonging high-benefit and -potential cylinder batch, increasing economic benefits and finally on the condition of not increasing production cost, increasing the total economic benefits of a whole workshop.