44results about How to "Automatic load balancing" patented technology

The present invention provides a cluster load balancing method transparent to an operation system. Main functional modules comprise a load balancing module, a processor migrating module and a communication module. The method is characterized by comprising the following steps: 1, driving a virtual processor to migrate; 2, driving balance migration; 3, sending a migrating request to a target node and negotiating; 4, storing and restoring a state of the virtual processor; and 5, communicating. The method better solves the problem of low resource utilization rate of a cluster system. Along with the development of more popularization of the cluster system and the continuous development of hardware virtualization technology in the future, the method can be a good solution for the low resource utilization rate of the cluster system and has good application prospect.

The invention relates to a server system, including a client end to initiate a login request of a game user; an account server store the account data of every game user in a game; a role service to store the role data requested by every game user in the game; in addition, many game server groups, every game server in each group runs the same game; and a game gateway server to verify the login request of the client end, after the client end is passed, allocate a role requested by the game user and determine a game server for running the game according to the role information and the load conditions of various game server in a corresponding game server group. Because it can implement the automatic load balance of various game servers by the game gateway server, it can support more on-line game users at the same time and add crossing servers to implement the role cross of a game user between different game servers, and has a wide application range.

The invention discloses a device and a method for balancing an active load of a Docker container. The method comprises the following steps: receiving a to-be-executed application file and presetting to-be-executed application details; converting the to-be-executed application file into a Docker mirror image; executing an application starting command in the Docker mirror image; transmitting a registration request and registering an application to a distributed coordination server; acquiring registering information from the distributed coordination server, acquiring change data, corresponding to the container, of an application example, and analyzing the data; updating a configuration file to a general load balancer by configuring a template. According to the device and the method disclosed by the invention, a real-time active discovering and registering mechanism is realized; the general load balancer is adopted for providing services for a Docker container cluster.

The invention discloses a distributed system and an automatic maintaining method for the same. The method comprises the following steps that after a node starts a service, a service registration request is sent to a service registry at regular time; and the service registry maintains local information according to the received service registration request and provides an information service for the node or a node related to the node. The scale of the distributed system disclosed by the invention can be dynamically shrunk; after the node starts the service, service registration can be automatically initiated without manual configuration; a network relation can be automatically formed between the nodes by a calling relation and can be dynamically regulated; and the network structure does not need to be planned in advance. Moreover, the system disclosed by the invention is simple to maintain and is low in construction cost.

The embodiment of the invention discloses a dynamic load balancing method and system for a container cluster. The method comprises the following steps: initiating connection to a load balancer by a container and registering; upon receiving a registration request from a newly added container, creating queue nodes corresponding to the service provided by the newly added container by the load balancer, mounting the created queue nodes to a queue corresponding to an identification mark of the service provided by the newly added container; when receiving a service request from a service request source, determining the queue corresponding to the service identification mark carried in the service request; determining the container corresponding to the node with minimum load in the determined queue as the target container; forwarding the service request to the target container. In the process, the balance configuration file is not required to be configured in advance, the load balancer is not required to restart, but the load balance is realized automatically, so that the dynamic load balancing method solves the problem that the existing load balancing method is difficult to adapt to the container cluster which is used for providing back-end service and changeable dynamically in a cloud computing system.

The embodiment of the invention discloses a distributed data computing method, device, computer equipment and a readable storage medium. The method comprises steps: event messages generated accordingto preset rules are sequentially stored in a message queue of a message middleware; according to a first preset rule, a plurality of message events in the message queue are read sequentially and are transmitted to an Ignite task manager; the Ignite task manager is controlled to count the number of idle threads in each application server in a server cluster, so as to determine the application server with the maximum idle threads as a target server, wherein the server cluster comprises a plurality of application servers, and each application server comprises a plurality of threads; and the Ignite task manager is controlled to transmit the plurality of received message events to the determined target server to perform corresponding data calculation. Computing tasks can be dynamically allocated, computing resources are ensured to be used more efficiently, and the most efficient computing performance is provided.

The invention discloses a streaming media scheduling and distribution method capable of reducing wait time of a user, aims at solving the technical problems of long network delay time, high disc fault rate and the like existing in the prior art, and provides the streaming media scheduling and distribution method capable of reducing wait time of the user in combination with a peer-to-peer (P2P) technique and the characteristics of a mobile internet. The streaming media scheduling and distribution method capable of reducing the wait time of the user comprises the following steps that: a plurality of areas form a distributed streaming media distribution system and a data center is arranged in the center of the areas; different files are dispersedly stored in a slice-free method; the areas communicate with each other through media routers to form a P2P network; each area comprises a media router and a plurality of media engines; resource pre-fetching operation is set; the media engines monitor system operating load in real time and report the relevant information of the media routers; and the media engines are connected and communicated with each other, media resources are freely scheduled among the engines and the like. The streaming media scheduling and distribution method capable of reducing the wait time of the user can be widely used for the on-demand and direct broadcasting service of the mobile internet.

The invention provides a big data technology-based electric power fee controlling method and system so as to solve problems of low storage capability, low calculation performance and poor system stability of a conventional electric power fee controlling system in big data quantity conditions. The big data technology-based electric power fee controlling method and system specifically comprises the following steps: 1, a big quantity of data is written in, searched and updated via distributed storage technologies; 2, a distributed internal storage calculation engine is used for data storage, verification and calculation in a whole fee calculation process; 3, a distributed file system, clustering resource scheduling and big data clusters are used as a basic framework, and physical nodes of the system can be dynamically expanded and removed. The big data technology-based electric power fee controlling method and system can provide mass data storage capability and high performance data calculating and analyzing capability, horizontal expansion capability of the fee controlling system can be increased, hardware cost can be lowered, fault tolerating capability of the system can be improved, fee control calculation stability can be improved, and application of big data technologies to electric power fee control can be forcefully supported.

The invention discloses a power-grid big time series data storage method. The information of a batch of measurement points having location relevance to each other in service logic is re-described by selecting an open-source distributed column-oriented database HBase as a storage layer, and combining an SG-CIM model in power grid service; a reasonable index organization mode of a measurement point data storage table is designed, and the physical storage positions of the historical data of the batch of measurement points having location relevance to each other in service logic are adjacent to each other by use of the partitioning and load balancing functions of the HBase, so that the disk seek time can be reduced for querying the historical data of the batch of measurement points, and the query efficiency can be improved; instant query service can be provided for service application.

The invention relates to an electricity changing waiting-time determining system and method, an electricity changing station, an electricity changing system and a recording medium. The electricity changing waiting-time determining system comprises a queuing sequence determining unit and an electricity changing waiting-time calculating unit, wherein the queuing sequence determining unit is configured to determine a queuing sequence number for a vehicle user who applies for electricity changing, and the electricity changing waiting-time calculating unit is configured to calculate a corresponding electricity changing waiting time based on the queuing sequence number determined by the queuing sequence determining unit.

The invention relates to a network function distributed elastic control method which employs a load estimation module to measure and assess control plane and data real-time loads, and predict load change according to historic loads, and employs a distributed decision-making device module to operate on a control plane, start an interchanger transfer strategy based on the assessment result of the load estimation module, determine the transfer scheme of programmable nodes, design the communication process of different control domains in a decision-making process, and store the state view of a whole network scope. The method can transfer programmable nodes in different control domains, realize the elastic control distribution of network functions, and allow networks to realize automatic load balance; the load estimation module operates on a controller, tracks and runs statistics on controller loads, and triggers the distributed decision-making device module to perform interchanger transfer; each controller in the distributed decision-making device module operates respective decision-making device instance, calculates an interchanger transfer strategy, and realizes maximum network resource effectiveness.

The invention discloses a method for controlling a task server and a system for controlling the task server. The method comprises the following step: when a central control server receives a triggering instruct, obtaining server information stored at local, wherein the server information includes current state of a task server and time for task implementation of the task server; according to the server information, judging if the current state of the task server is an idle state; if it is, determining that time for implementing task belongs to the first preset time scale in the task server under the idle state, and customizing it as a first set; distributing the task required to execute to the task server in the first set. Through above method, the method and system can automatically balance the load of the task server, and thereby resources are effectively utilized and the service quality is guaranteed.

An embodiment of the invention discloses a relational database service system. The system comprises an access service processing device, a manager and a monitoring execution processing device, wherein the access service processing device provides a consistency database access interface, keeps a transparent data access service to an application, carries out equilibrium assignment on the access service of a user to a database according to routing information obtained from the monitoring execution processing device and sends the access service to the manager; the manager provides a unified cluster management framework and sends an operating instruction to the monitoring execution processing device according to the obtained access service; the monitoring execution processing device obtains first information when the corresponding service on a host where the monitoring execution processing device is located runs, carries out monitoring processing on the first information and responds to the operating instruction sent by the manager to manage the database.

The invention relates to an ultra-high pressure balanced load balancing valve and a method thereof. The ultra-high pressure balanced load balancing valve comprises a valve cavity, wherein the valve cavity is formed in a valve body and an upper part and a lower part of the valve cavity communicates with the outer part of the valve cavity; a valve core is movably arranged in a valve sleeve; an uppersealing base and a lower sealing base are respectively arranged at the two ends of the valve core in a sleeving manner; a pressure adjustment spring device and a conical valve seat spring device arerespectively arranged at the upper end and the lower end of the valve cavity in a sealed manner; an upper hydraulic chamber is arranged between the valve core and the upper sealing base and communicates with an upper oil hole; a lower hydraulic chamber communicates with a lower oil hole; the upper hydraulic chamber communicates with the lower hydraulic chamber through a channel arranged in the valve core; the valve sleeve is also provided with an intermediate hydraulic chamber formed in the middle end of the valve core; and the intermediate hydraulic chamber is connected with a pilot control hole and an oil passing hole. The ultra-high pressure balanced load balancing valve disclosed by the invention is mounted on an oil cylinder, the load balancing valve can be automatically turned on when the pressure in the oil cylinder exceeds the regulated pressure, over-high oil pressure is unloaded, a function of balancing load is achieved when one load is synchronously jacked by a plurality ofoil cylinders, and the oil cylinder is protected from being overloaded and damage on the oil cylinder is avoided, and therefore, the service life of the oil cylinder is prolonged, and the economical efficiency is promoted.

The invention discloses a dynamic service function chain arrangement method and system based on deep reinforcement learning. The method comprises the following steps: acquiring a historical network state according to an SDN controller; the network state comprises service function chain request stream information generated in the Internet of Things supporting mobile edge computing and corresponding network resource state information; setting deep reinforcement learning parameters and initializing the weight of the neural network; training a neural network according to an experience sample generated by interaction of the intelligent agent and the environment; for the service function chain request flow obtained in real time, the trained neural network is utilized, a heuristic algorithm is adopted, the placement and routing path of the virtualized network function meeting the requirement of the service function chain request flow is determined and deployed, and network resource state information is comprehensively considered; the load balancing of the network is realized while the resource consumption cost and the time delay of the request stream of the Internet of Things are reduced, and the network flow receiving rate is improved.

The invention discloses a Gigabit and 10-Gigabit Ethernet intercommunication system and method. The Gigabit and 10-Gigabit Ethernet intercommunication system comprises a Gigabit Ethernet transceiver interface module, a 10-Gigabit Ethernet transceiver interface module and a logic module. The logic module is configured between the Gigabit Ethernet transceiver interface module and the 10-Gigabit Ethernet transceiver interface module, and is used for gathering Gigabit Ethernet traffic from the Gigabit Ethernet transceiver interface module to the 10-Gigabit Ethernet transceiver interface module and outputting the Gigabit Ethernet traffic, and equally allocating 10-Gigabit Ethernet traffic from the 10-Gigabit Ethernet transceiver interface module to the Gigabit Ethernet transceiver interface module and outputting the 10-Gigabit Ethernet traffic. Compared with the prior art, the Gigabit and 10-Gigabit Ethernet intercommunication system and method solve a problem that traffic cannot get access to a network device because of incompatibility with a 10-Gigabit Gigabit interface on a selected network hardware device in a Gigabit or 10-Gigabit Ethernet traffic environment, and are reasonable in design, simple, feasible, good in practicality.

The invention discloses a flexible self-balance multi-suspension-point hanging beam structure. The flexible self-balance multi-suspension-point hanging beam structure comprises a crossbeam of a box structure and a steel wire rope, wherein steel wire rope fixing ends are arranged in a cavity of the crossbeam, fixed pulleys are arranged among the steel wire rope fixed ends, and the flexible self-balance multi-suspension-point hanging beam structure also comprises a plurality of riggings; the two steel wire rope fixing ends are respectively provided with a steel wire rope buckling position and a fixed end wheel disc, each fixed pulley and each movable pulley are respectively provided with a pulley wheel disc which is arranged side by side, one end of the steel wire rope is fixedly installed on one steel wire rope buckling position, and the other end of the steel wire rope sequentially bypasses the pulley wheel discs and the two fixed end wheel discs to continuously bypass the pulley wheel discs so as to be installed on the other steel wire rope buckling position. By adopting the structure, the load of each suspension point can be automatically balanced, so that the hanging beam is stable and reliable in work, the structure is simple, and convenience in manufacturing is realized; multiple operating conditions can be realized by virtue of different combinations of the suspension points, various working requirements can be met, and the utilization rate and the working efficiency of the hanging beam can be improved.

The invention discloses a system for realizing automatic load balancing of openstack swift service. The system comprises a client, a service registration and authentication node, a load balancing plug-in module, an acquisition monitoring module, a storage service gateway agent node and a storage node, wherein the load balancing plug-in module and the acquisition monitoring module form a control link; the load balancing plug-in module is used for modifying and setting a terminal address of a distributed object storage service in the service registration and authentication node or a terminal address of a server object storage service in the service registration and authentication node according to a balancing rule and the monitoring state information of the acquisition monitoring module; andthe acquisition monitoring module is used for monitoring the state of the storage service gateway proxy node server and feeding back the monitored state information to the load balancing plug-in module of the service registration and authentication node. The system and the method have the advantages that the openstack swift service automatic load balancing is realized, the equipment utilization efficiency is improved, and the maintenance cost is greatly saved.

The invention discloses a decentralized HPC computing cluster management method and system based on a paxos algorithm. The method comprises the steps of deploying a main management node and a plurality of standby management nodes, and setting a cluster management election mechanism, wherein the cluster management election mechanism comprises the steps that a reply of heartbeat connection sent by amain management node exceeds a preset value, and a standby management node carries out election according to the paxos algorithm to generate a new main management node; and enabling the original mainmanagement node to be offline, and enabling the new main management node to perform heartbeat monitoring on the remaining standby management nodes. According to the invention, the HPC high-performance job scheduling cluster can be optimized from a single-master centralized cluster mode to a decentralized cluster mode; the change of the mode enables the availability of the cluster to be greatly improved; limit of a single-point fault of the single-master cluster centralization mode is avoided; the fault-tolerant capability of the cluster is improved by several orders of magnitudes; the faultis more suitable for an actual scene; automatic high availability is provided for the cluster; and high availability is completed without a third-party tool.

Taking full advantage of processing power of control device, the disclosed method reduces number of layer to handle data transmission and shortens I / O route so as to overcome disadvantages in access speed and control of existing network storage technique. The invention includes interface of memory device, object interface for network storage. Interface of memory device executes protocol oriented to block device level. Through network interface, user accesses storage objects, and is able to carry out pointed operations for objects. The method runs automatically if condition is met, or runs when an event is triggered by user's request. Improving traditional data management and processing mode, the invention uses OBSC to replace data management under bottom of file system and database system as well as part of data process taken by application program at client end. The method is easy to realize function of self-management such as self-recovery, load balancing and hotspot movement.

The invention provides a load balancing method based on an XMPP communication processing server. The specific method comprises the following steps: a terminal agent sends a request to a control centerto obtain its own communication account number; the terminal agent sends a request to the XMPP server openfire to obtain a list of possible terminal servers in the network; XMPP communication is usedto send load inquiry protocol to all the terminal servers in the list of terminal servers; the terminal server with the lowest load is selected from the list of terminal servers that return the loadresponse; a service login protocol is initiated to the terminal server with the lowest load, and the address / port of the terminal server with the lowest load is obtained; The communication account ofthe terminal server carries the network information and the server equipment information to distinguish the network where the terminal server is located from the different server equipment in the sameLAN. Compared with the prior art, on the basis of the simple push-pull combination mode, multi-network deployment and automatic load balancing are supported.

The invention discloses a load balancing method and device, equipment and a storage medium, and relates to the field of edge cloud computing. The method comprises the following steps: acquiring resource use information of a current edge cloud; According to the resource use information of the current edge cloud, determining whether cross-edge cloud automatic load balancing needs to be carried out or not; if it is determined that cross-edge cloud automatic load balancing needs to be carried out, selecting a new edge cloud superior to the current edge cloud, and accessing a terminal to be accessed to the current edge cloud or a terminal which is accessed to the current edge cloud to carry out service processing to the new edge cloud. According to the technical scheme, automatic access switching and load balancing among multiple edge clouds can be achieved, the resource utilization rate is increased, and resource waste is reduced.

The embodiment of the invention provides an automatic load balancing method and device of a distributed system, which can be used in the technical field of cloud computing, and the method comprises the following steps: based on preset automatic load balancing parameters, selecting a source computing node and a target computing node to be subjected to load balancing in the current distributed system, selecting a target virtual machine from the virtual machines corresponding to the source computing node; and if the target computing node meets the resource setting requirement of the target virtual machine, migrating the target virtual machine from the source computing node to the target computing node. According to the load balancing method and device, load balancing between the computing nodes in the distributed system can be automatically achieved, the processing timeliness and reliability of the load balancing process can be effectively improved, the accuracy and effectiveness of a load balancing result can be effectively improved, and then the resource utilization rate and operation stability of the distributed system can be effectively improved.

The invention relates to a system and method for determining waiting time for battery swap, a battery swap station, a battery swap system, and a recording medium. Wherein, the battery change waiting time determination system includes: a queuing sequence determination unit configured to determine a sequence number for a vehicle user applying for a battery change; and a battery change waiting time calculation unit configured to The sequence number determined by the determination unit is used to calculate the corresponding battery replacement waiting time.

The invention discloses a GPU resource pool scheduling system, which includes a GPU cloud computing power center and a GPU cloud control node. The GPU cloud computing power center includes a plurality of GPU computing power units, and each GPU computing power unit includes VMM and RC. The control node includes RS; the GPU computing power unit is used to provide GPU computing power; VMM is used to provide a control interface and receive resource scheduling instructions from RS, create a virtual machine according to the instructions, allocate vGPU resources to the virtual machine and start the virtual machine; RC It is used to count the resource data of the GPU computing power unit and report it to RS; RS is used to collect the resource data reported by each RC, and send resource scheduling instructions to each VMM, and globally schedule the resources of the GPU computing power unit, including the allocation of GPU resources. Gather to form multiple groups of GPU hardware collections, and multiple groups of GPU hardware collections form a GPU resource pool; the system realizes the unified resource pool scheduling management of multiple manufacturers, GPU models, and GPU virtualization methods in the cloud computing platform; the invention also discloses A GPU resource pool scheduling method is proposed.

An embodiment of the invention discloses a relational database service system. The system comprises an access service processing device, a manager and a monitoring execution processing device, wherein the access service processing device provides a consistency database access interface, keeps a transparent data access service to an application, carries out equilibrium assignment on the access service of a user to a database according to routing information obtained from the monitoring execution processing device and sends the access service to the manager; the manager provides a unified cluster management framework and sends an operating instruction to the monitoring execution processing device according to the obtained access service; the monitoring execution processing device obtains first information when the corresponding service on a host where the monitoring execution processing device is located runs, carries out monitoring processing on the first information and responds to the operating instruction sent by the manager to manage the database.

The invention provides a load balancing method based on an XMPP communication processing server. The specific method includes: the terminal agent sends a request to the control center to obtain its own communication account; sends a request to the XMPP server openfire to obtain a list of terminal servers that may exist in the network ; Use XMPP communication to send a load inquiry protocol to all terminal servers in the terminal server list; select the terminal server with the lowest load from the terminal server list that returns the load response; initiate a business login protocol to the terminal server with the lowest load to obtain the lowest load The address / port of the terminal server; wherein, the communication account of the terminal server carries the network information and server device information to distinguish the network where the terminal server is located and different server devices in the same local area network. Compared with the existing technology, based on the simple push + pull combination mode, it realizes support for multi-network deployment and automatic load balancing.