111 results about "Autoscaling" patented technology

The invention discloses a resource management system and method for containerized application, wherein the system is composed of a management node and at least one computing node. The management node provides application management and resource monitoring function, and schedules tasks on demand according to the resources of the application demand and the resources that the computing node can provide, and also monitors the resource utilization ratio of the CPU, I / O devices and network devices of each computing node to guide the application scheduling to perform accurately. A container state analyzer in the calculation node performs the resource consumption statistics and analysis to the currently running container. Whether the running container produces the performance disturbance is distinguished and quantitative analyzed based on an SVM interference detection module. When performing performance disturbance, a container resource dispatch module dynamically adjusts all resources of the container. Efficient deployment and scheduling scheme are provided for containerized application in the cloud environment. The computing resources can be adjusted according to the dynamic changes of the applied workloads, and the elastic expansion and contraction are adaptive.

The present invention provides an elastic expansion method based on a container cluster and an application system thereof. The method comprises the steps of (1) creating a container cluster, and setting the threshold of elastic expansion, (2) judging whether a container in the container cluster is started or not, and executing a corresponding operation according to a judgment result, (3) setting a unified access point when all containers are started, (4) triggering elastic expansion when the indicator of the container in the container cluster reaches the set threshold, and adding or deleting the container, (5) in the elastic expansion process, adjusting the unified access point as a load balance strategy, which means that an access flow is always oriented to a viable container, (6) adding a new container into the range managed by the unified access point if the new container exists after the elastic expansion is completed. According to the method and the system, the number of the container cluster can be dynamically adjusted according to a monitoring data result, thus online application access in not influenced by application traffic, the maximization of resource utilization is realized, and the waste of resources in an idle time is avoided.

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 host node election method and nodes. The method includes the following steps: each node of a cluster requests and sends proposals to all the other nodes in the cluster, judging whether amount of nodes sending proposals on their own reaches the required minimum amount allowing proposal sending, if the result is yes, broadcasting the proposals to all the other nodes in the cluster, judging whether the amount of nodes sending proposals on their own reaches the required minimum amount with accepted proposals, if the result is no, deciding the proposal node to be the host node, and broadcasting the accepted proposals to all the other nodes in the cluster. The host node election method and the nodes select the host node from the nodes in the cluster in a dynamic negotiation mode, and decide the host node under the condition of elastic scaling of the amount of the nodes in the cluster.

The invention discloses a method, apparatus and system for realizing elastic expansion of an application example based on a container. The method comprises the following steps: determining a target host node and a target container corresponding to a target application; determining an index of an elastic expansion strategy corresponding to the target application; obtaining an actual value of the index corresponding to the target container; and if it is judged that the number of the containers of the target application needs to be adjusted according to the actual value of the index and the elastic expansion strategy, sending a container number adjustment instruction to a Docker Swarm component, so that the Docker Swarm component adjusts the number of the containers. Therefore, in the embodiment, the obtained actual value of the index is matched with the predetermined elastic expansion strategy to automatically trigger the Swarm cluster component to perform scheduling to create or delete the containers, automatic capacity expansion or expansion of the application example under a certain condition, thereby improving the high reliability of the application based on the container.

The invention relates to the technical field of cloud computing and especially relates to an OnAir all-media cloud platform application method. The method comprises the following steps: 1) user demand submitting: a user submits a required service to an OnAir all-media cloud platform through a platform interface; 2) OnAir all-media cloud platform data processing: the OnAir all-media cloud platform provides the corresponding service according to the user demand, and carries out service lifecycle management; and 3) OnAir all-media cloud platform data management: the OnAir all-media cloud platform manages user and service data through a business service system and a cloud service engine, and coordinates resources of different cloud service providers through the cloud service engine and middleware and carries out data management. Through the method, an OnAir all-media cloud platform application is good in opening and compatibility, and realizes service support elastic expansion, short service provisioning period and high service processing efficiency.

The invention provides a democratic and autonomous cluster management method. The democratic and autonomous cluster management method includes the following steps of: step 1, cluster startup; step 2, cluster operation autonomy; step 3, cluster split autonomy; step 4, cluster merged autonomy; step 5, cluster dynamic elastic autonomy; and step 6, cluster administrator governance. The invention also provides a corresponding system. With the democratic and autonomous cluster management method and system of the invention adopted, a cluster can have continuous external service and cluster management service providing ability when any single node is abnormal or a plurality of nodes are abnormal; at any state, maintenance upgrade can be performed on nodes in the cluster, and service nodes can be added into the cluster; third-party services are not required as for the selection of a master; and democratic and autonomous selection and generation of the master can be competed; split autonomy of the cluster is supported; fault tolerance and management performance can be improved; the highest administrative authority of the administrator is supported. The democratic and autonomous cluster management method can be widely applied to cluster management systems which require elastic flexibility, have high requirements for continuous service quality and have a large number of nodes in clusters.

The invention discloses an elastic scalability method and system, and belongs to the field of communications. The method comprises the following steps: acquiring monitoring data of an application supported by a scalability pool, wherein the scalability pool comprises at least one scalability group, each scalability group comprises at least one pre-created virtual machine, and each virtual machine is pre-deployed with an application; performing scalability decision on each scalability group according to the monitoring data; when deciding to expand the virtual machine in the target scalability group, selecting the first virtual machine from the virtual machines in dormancy state included by the target scalability group, and awakening the first virtual machine. Through the adoption of the method disclosed by the invention, the problem that the elastic scalability scheme provided by the prior art cannot meet the fast expansion demand; through the pre-creating of the virtual machine and the pre-deploying of the application, the virtual machine in the dormancy state is awakened, namely, the fast expansion of the application node is realized, the technical effects of improving the scalability speed and meeting the fast scalability demand are achieved.

An apparatus comprising a method for allocating shared processing resources, comprising the steps of (A) predicting a first amount of processing resources needed to complete each of a plurality of transcoding jobs in a work queue of a project, (B) determining a second amount of processing resources available prior to processing the project and (C) adding processing resources needed to complete the project. The processing resources are added in a fixed number of predetermined processing units. The fixed number of computer resources are added (a) to minimize an overall use of said processing resources needed to complete the project and (b) to complete the project within a target time.

The invention discloses a virtual machine processor resource adjusting method and device and a virtual machine system. The method comprises, for a VCPU (virtual central processor unit) to be deleted of a virtual machine (VM), identifying whether the VCPU contains a resource scheduling request; responding to the resource scheduling request of the VCPU, according to the response time threshold value of the operating system of the VM for starting automatic fault detecting operation, delaying the response to the resource scheduling request, and enabling the resource scheduling request to obtain response within the response time threshold value; performing response when the resource scheduling request does not obtain the response within the response time threshold value, starting the automatic fault detecting operation through the operating system of the VM, scheduling queued processes in the VCPU to other VCPUs of the VM; deleting the VCPU, and releasing physical server CPU (central processor unit) resources occupied by the VCPU. The virtual machine processor resource adjusting method can reduce VCPU resources in an operating state and achieve flexible expansion and retraction of virtual machine resources.

The invention discloses a method and device for an elastic extensible virtual machine cluster. A specific embodiment of the method includes: when arranging a service, calculating first left resource quantity of virtual machines in the virtual machine cluster; acquiring resource demand of the to-be-arranged service; acquiring configuration grade of the virtual machines in the virtual machine cluster; according to whether the first left resource quantity meets the resource demand or not, determining a virtual machine with a preset configuration grade as a target virtual machine; arranging the to-be-arranged service to the target virtual machine. By the embodiment, the service can be arranged in the virtual machine cluster including different configuration grades, and different services can be arranged on a single virtual machine in a mixed manner, so that service arranging cost is lowered, and resource utilization rate of the virtual machine cluster is increased.

The invention discloses a cloud resource scheduling method. The cloud resource scheduling method and device comprises the steps that a resource scheduling model is generated on the basis of an elastic stretching strategy and a dynamic migration strategy by combining the scale of a cloud platform; transverse and / or longitudinal multi-dimensional scheduling is conducted on virtual resources of which resource allocation exceeds a preset value through the elastic stretching strategy and the dynamic migration strategy in the resource scheduling model. According to the cloud resource scheduling method, by adopting the resource scheduling model which is generated on the basis of the elastic stretching strategy and the dynamic migration strategy by combing the scale of the cloud platform to conduct dynamic scheduling on the virtual resources in the platform, the migration and bad dynamic loading effect problems which are likely to be caused by an instant resource utilization rate peak value can be avoided, the resource using efficiency of the cloud platform is effectively improved, the stability of the cloud platform is improved, the cloud platform is safer and more reliable, and the user experience degree is increased. The invention further discloses a cloud resource scheduling device which also has the above-mentioned advantages.

The invention discloses a cluster top-speed elastic telescoping method for realizing efficient utilization of resources. An intelligent telescoping compensation process comprises the following steps:1, creating an intelligent elastic telescoping compensation module IACM and a compensation queue; 2, judging whether the service cluster needs to add nodes or not in a future period of time; 3, addingthe Pod resources into a compensation queue; 4, setting the priority of the compensation queue; 5, carrying out recovery operation on the Pod resources; 6, maintaining Pod resources; and 7, periodically judging. According to the cluster top-speed elastic scaling method, when the resource demand of the service cluster is increased, the new copy already exists in the compensation queue and can be added into the cluster by directly pulling up the new copy, so that the time for establishing the new copy is saved, and the target of more extremely and quickly expanding the cluster node resources isachieved. When resources in the resource pool are deficient, resource adjustment is performed, the resource proportion is approximately unadjustable, and expelling or killing tasks can be performed to realize resource scheduling and resource scaling.

The embodiment of the invention discloses an elastic scaling method and a device for a cloud service. One specific embodiment of the method comprises the following steps: acquiring monitoring information of at least one resource in a cloud service system; matching the monitoring information with a preset scaling rule; if the expansion condition is met, selecting resources for expansion from a predetermined expansion group according to an expansion rule; and installation the resources for capacity expansion to the load balancing equipment or directly accessing the public network IP according toa preset load balancing mode. According to the embodiment, automatic adjustment of cloud service resources in various scenes is realized.

The invention provides a virtualized container service elastic scaling method oriented to data center load monitoring. Within the cloud platform cluster, at first, the real-time load of the containerservice is obtained through the load monitor service, and according to the load and the relationship of the upper bound and the lower bound of the set threshold, whether the elastic scaling of the container service is triggered is determined, after detecting that the container load pressure exceeds the upper limit of the threshold, container service expansion is triggered. By synthetically scoringdifferent nodes in the cluster with various algorithms, the best adapter node for elastic expansion and contraction is obtained, and finally a new container is started on the adapter node to completethe container service expansion. As the load pressure of the container is detected to be low than the lower threshold limit, the container service shrinkage is triggered, and elastic expansion and contraction of the container service are completed by removing the corresponding container on the node marked in the expansion phase. The method can realize the elastic expansion of cluster inner container service on the premise of keeping the load balance of cluster resources.

Embodiments of the present invention provide a method and an apparatus for automatically and flexibly expanding and contracting virtual resources. The method comprises: receiving a monitoring alarm for at least one working indicator of a target expanding and contracting group; determining an expanding and contracting policy for the target expanding and contracting group according to the monitoringalarm; when the expanding and contracting policy is adding an instance, obtaining the instance specification of the target expanding and contracting group, wherein the instance specification is determined in advance according to the statistical information of the monitoring data of at least one working indicator of the target expanding and contracting group; and sending an adding instance requestcarrying the instance specification to an instance resource management platform, so that the instance resource management platform adds an instance to the target expanding and contracting group according to the instance specification. The instance specification of the added instance in the embodiments of the present invention is updated according to the load conditions reflected by the working indicators of the expanding and contracting group, and resource allocation accuracy and resource utilization of instances of the expanding and contracting group can be improved.

The invention discloses a lightweight concurrent migration method of a database, and relates to the technical field of data migration. Aiming at problems and existing defects in the existing database migration process, the technical scheme adopted by the invention comprises the following steps: loading configurations of an original database and a target database, loading a corresponding grammar dialect according to the type of the target database, and performing grammar verification on the configurations; generating a table, a storage process, a function, a view, an index and others of the target database based on the loaded configuration; according to resources of an operating system, dynamically controlling the number of started threads, and carrying out concurrent migration on data of an original database through the started threads; and after migration is completed, performing default verification and configuration verification on the original database and the target database. Third-party middleware is not adopted, the method has the advantages of being efficient in operation, low in use cost, easy to expand, configurable and capable of achieving elastic expansion and contraction, and lightweight fast migration of data can be achieved.

The invention particularly relates to a Heat-based cloud application deployment scheme. According to the Heat-based cloud application deployment scheme, Heat and Agent are well combined together; the Heat is in charge of application running environment creation, capacity pre-configuration, elastic stretching, load balancing and virtual machine layer monitoring; and the Agent is in charge of application publishing and updating, log acquisition and application life cycle monitoring. The Heat-based cloud application deployment scheme has the advantages that the application deployment is simpler, more convenient, more intelligent and easier to manage, the operation and maintenance costs and the personnel cost are greatly reduced, and the deployment cycle is shortened. Each application can run different quantities of instances according to various emergency situations, so that the resource overhead cost is reduced and the application running stability is ensured.

The invention discloses an education service architecture system based on a container cluster. The system comprises a physical resource layer, an operating system layer, a resource management layer and a teaching application layer which are connected in sequence from bottom to top, all the layers carry out cooperation and communication management through a distributed coordination manager Zookeeper; the teaching application layer provides a server interface to provide teaching service application for an external system. Kubernetes serve as a framework and are integrated into the resource management layer to obtain bottom-layer physical resources to manage teaching application services in the container cluster; wherein the container cluster is used for running a Docker application, and theexternal system calls the teaching application service in the architecture system through the teaching application layer. According to the education service architecture system based on the containercluster, rapid deployment can be realized, various service applications and public service platforms of schools can be stopped at a second level, and cluster deployment, load balancing, disaster recovery and elastic expansion and contraction of various services are supported.

The invention discloses an automatic deployment method and device based on helm, and the method comprises the steps: employing helm as a deployment template file of Kubernetes, and employing Velocityrendering to generate a charts file of helm; calling a helm Api file and a charts file to generate a docker mirror image, and arranging and scheduling the docker mirror image into a Kubernetes cluster; according to the specified resource index data, analyzing the system load condition through the log, the throughput and the system operation condition, and carrying out elastic expansion and contraction. The method and the device are based on a helm, so that application deployment of the Internet of Things can be completed, and page query deployment information can be provided; elastic expansionand contraction are supported, and dynamic expansion and contraction are carried out according to resources used for analyzing the current application and a set threshold value; custom ip or domain name access is supported; persistent storage of application mirror images, logs and configuration files is supported.

The invention discloses a container-based application management method and device. A server-less computing system implemented based on a container is configured such that an application instance is allowed to be in one of an online state and a low power consumption state at runtime. In response to the execution of the capacity reduction processing on the application, a first application instance of at least one online state of the application is enabled to enter a low-power-consumption state; and in response to execution of the capacity expansion processing on the application, at least one second application instance in the low-power-consumption state of the application is enabled to enter an online state. Therefore, the cost of the application instance can be reduced, and the application instance can be quickly and elastically expanded and contracted.

The invention discloses an intelligent elastic telescoping method based on a container cluster. The method comprises the following steps: creating a container cluster, and setting a threshold value ofelastic expansion, the threshold value being the maximum CPU usage rate, the maximum memory usage rate, the maximum disk IO, the maximum network throughput, the longest request response time and themaximum request rate of a target time period when a container runs; judging whether the containers in the container cluster are started or not, and executing corresponding operation according to a judgment result; after all the containers are started, setting a unified access point; enabling the monitoring module to monitor the running state of each host; sending the data acquired at regular timeto a monitoring server; enabling the monitoring module on each host to be responsible for sampling, counting and processing the real-time and historical CPU usage rate, memory usage rate, disk IO, network throughput, request response time, request rate and the like of the host and the container, the method can be flexibly configured and created, the resource occupation rate is low, and the telescoping method is more flexible and efficient.

The invention relates to an elastic scaling method based on load prediction and a system. The elastic scaling method comprises the steps of determining the current service request data according to application load data within a preset historical time range and a first preset rule; generating a corresponding scaling rule when the current service request data meets preset scaling requirements so asto trigger a scaling activity request; creating a scaling activity according to the scaling activity request; and executing the scaling activity so as to add or delete a cloud server instance of a scaling group. The elastic scaling method can provide elastic services timely and effectively, realizes on-demand supply of resources and can better adapt to application scenarios of a large-scale cluster.

The invention discloses a strategy-based cloud database elastic scaling method, which comprises the following steps of: setting a database upper limit value contained in a single cloud server, and setting a data table upper limit value contained in a database; Setting an elastic scaling strategy hit by each cloud server based on the feature data; Dividing the data into an expansion or shrinkage target database and a data table based on the elastic scaling strategy. According to business development requirements, the capacity expansion or reduction of the database is automatically and flexiblyadjusted and completed, the business application requirement can be well met, resources are well saved by reducing the capacity of the database in time, the business application service does not needto be stopped, the capacity expansion or reduction work is completed under the non-perceptual condition, and great convenience is brought to a user to use.

The invention discloses a method for elastic scalability configuration in a cloud computing environment. The method comprises the following steps: acquiring current load information of at least one performance indicator dimension in the cloud computing environment; determining whether the load information exceeds a pre-configured load threshold interval, and if so, determining a corresponding loadcurve by using a load information set within the specific time interval in the cloud computing environment, and calculating load change information of the load information contained in the load information set in the specific time interval; and judging whether the load change information is within a pre-configured load change threshold range, and if so, performing elastic scalability configuration on cloud computing resources in the cloud computing environment on the performance indicator dimension. By adoption of the method for elastic scalability configuration in the cloud computing environment, the load information with changes excluding the preset change range in the cloud computing environment can be filtered to save the computing resources or improve the computational efficiency andimproving the computational stability.

The invention discloses an elastic scaling method based on a Kubernetes cluster, is applied to the technical field of computers, and aims to overcome the defects of an elastic scaling strategy built in Kubernetes, predicts future workload according to historical index values of the cluster on the basis of a deep learning platform. The absolute value of the difference between the current index value and the predicted index value is compared with the set threshold value, so that the condition of frequent capacity expansion and shrinkage of the service in the cluster can be avoided; if the absolute value is larger than or equal to a set threshold value, it is considered that an index peak or an index valley is about to occur in the future, the service in the cluster is necessary to execute capacity expansion or capacity shrinkage operation, and meanwhile, a timer is further arranged for recording time consumption of one time of capacity expansion or capacity shrinkage operation; in order to avoid the conflict between the current capacity expansion and contraction operation and the previous capacity expansion and contraction operation, a time threshold value is also set, and the capacity expansion or contraction operation is carried out only when the value of the timer is greater than or equal to the time threshold value.

The invention discloses an elastic scaling method based on load regression prediction in a cloud environment and a readable storage medium, belongs to the field of elastic scaling of computing resources, and aims to solve the technical problem of how to predict the user load in the next time period and reasonably adjust the computing resources of a user according to the load. The method comprisesthe following steps: acquiring historical data of normal operation of a user service in a period of time, and constructing a training data set; based on the training data set, training a regression model, calculating the load of the next moment through the trained regression model, training a linear model between the calculation resource usage amount and the load, and calculating the calculation resource usage amount of the next moment through the trained linear model; and optimizing the user cost by taking the computing resource usage amount at the next moment as a limiting condition to obtain a computing resource configuration target solution set, rearranging the computing resource configuration target solution set and the computing resource configuration current solution set, calculating to obtain a difference, and selecting a proper horizontal telescopic mode or vertical telescopic mode based on the difference.

The invention provides an instance elastic telescoping method for guaranteeing the stability of virtualization systems. The method provided by the invention comprises following steps: determining whether the load status of the instance in the telescoping group exceeds a first upper limit value; if the load status exceeds the first upper limit value, creating a monotonically increasing time window. In the monotonically increasing time window, determining whether the load status of the instance in the telescoping group exceeds a second upper limit value; if the load status exceeds the second upper limit value, adding n instances in the telescoping group. The invention also provides a related instance management module and a computer device.

The invention relates to a method, a device, a storage medium and an electronic device for elastically scaling an application program, which are applied to a cloud server. The method comprises: aftera monitoring device determines a policy identifier according to a first resource access data of the application program, the policy identifier sent by the monitoring device is received; a first elastic scaling policy is determined corresponding to the policy identifier; a corresponding elastic stretching operation is performed according to the first elastic stretching strategy.

The invention provides a network-function-virtualization (NFV) network element configuration method and system. A corresponding relation between meta information and super meta information of base station types under various core network elements is established; correspondence between meta information and super meta information of various manufactures and base stations with different types is carried out; when a core network element is generated in a virtualized mode, the correspondence relation is queried according to the type of the core network element and a base station type of a to-be-configured base station, so that meta information of the to-be-configured base station is obtained; and thus a configuration file that can be identified by the to-be-configured base station is generated and is sent to the to-be-configured base station, so that automatic configuration of the base station is realized. Therefore, a configuration server can generate a configuration file of a to-be-configured base station with a needed type automatically without hand-operated configuration by the technical staff of the corresponding manufacturer, so that the labor cost is saved, the access capability of the base station equipment is improved, and the automatic demand of elastic capability expansion of the NFV system is satisfied.