43results about How to "Avoid Overloaded Situations" patented technology

The invention discloses a Web request load balancing method based on operating system virtualization. The present invention firstly collects and normalizes server resource information; secondly collects server and service instance dynamic information; then calculates the final weights of service instance copies of all services according to server resource information and periodically collected dynamic information; Poll for distribution requests. The present invention can effectively avoid different service instance copies on the same server, and the server is overloaded due to the low load of the server receiving a large number of Web requests at the same time, realizes the load balance between the service instance copies and the servers, and improves the concurrency of the Web cluster system ability to reduce the average response time of requests.

The present invention discloses an adaptive and stable clustering method and system suitable for load balancing of highly dynamic wireless networks, wherein the method includes the following steps: S1. Calculate the cluster head competence of nodes according to the stable neighbor nodes, and thus Elect the optimal cluster head to form the initial cluster structure; S2. Adaptively maintain the cluster structure, including the maintenance of stable neighbor nodes, the storage and update of the standby follower node table, the change of cluster type, cluster fusion, cluster dissolution and multiple Maintenance of gateway node pairs. The present invention provides an adaptive and stable clustering method suitable for load balancing of highly dynamic wireless networks. Considering the impact of the clustering method on the three factors of clustering network connectivity, stability, and load balance, it can Density establishes a general and stable clustering structure in which multiple types of clusters coexist, making the clustering method applicable to various scenarios of large-scale traffic densities with uneven distribution of nodes.

Disclosed are a method and system for a terminal to access a network. The method includes: determining an access delay time for a terminal to re-access the RAN side when the terminal receives an RRC connection reject message or a connection release message sent from the RAN side; and initiating by the terminal an access request toward the RAN side when the waiting time of the terminal is equal to or exceeds the access delay time. By way of the present invention, the following problem in the related art can be solved: the requests of all the current MTC terminals will be rejected when congestion occurs in the core network entity, and if these MTC terminals simultaneously initiate access again subsequently, this will inevitably cause congestion which cannot be resolved.

The invention relates to an optimization control method of a transformer load of a transformer substation automation system, belonging to the technical field of transformer substation automation. The method comprises the steps of: acquiring a load related to running main transformer equipment by carrying out the topology connection analysis on the current transformer substation; if the main transformer equipment overloads, calculating a resectable load of the overloaded main transformer equipment according to a load priority and a locking state and submitting to a scheduling decision through a network; and practically issuing a resection control instruction in case of scheduling approval, or else recalculating a load resecting strategy according to a scheduling instruction result and submitting to the scheduling to be confirmed until the main transformer equipment has no resectable load any more. On a basis of an action strategy of a traditional transformer load optimization device, the invention frees the cost of a hardware device by embedding a software module in a transformer substation comprehensive monitoring system and interacts with a load optimization control strategy at amain station terminal by combining with the current computer communication technologies so as to achieve the aim of optimally controlling the transformer substation load.

The invention discloses a method for measuring true stress of a material and belongs to the technical field of material science. According to the method disclosed by the invention, a Poisson effect and volume expansion of the material in an elastic deformation region are considered; meanwhile, height localization of deformation of the material in a plastic deformation process is considered, and plastic deformation is finished through formation and sliding of a shear zone. Therefore, in a whole deformation process, an effective loading area can be accurately calculated by combining an originalsize of a sample piece, and the true stress on the sample piece can be accurately measured by utilizing a uniaxial loading load-displacement curve; the method has great significance and value in a process of determining basic mechanical property parameters by utilizing uniaxial loading.

The invention discloses an SDN-based multi-controller dynamic deployment method, comprising: using the Paxos algorithm to select a controller as the master node from a plurality of controllers adjacent to the faulty controller in the SDN network, and the master node according to the SDN network The topology structure of each follower node, the load of each follower node, and the message sent by the faulty controller during synchronization with it obtain the final migrating controller corresponding to all the switches managed by the faulty controller, and then the follower node downloads the faulty controller according to the obtained faulty controller. The final migration controllers corresponding to all the switches execute the migration process of the switches. The present invention can solve the problems in the existing multi-controller dynamic deployment method that the average response time delay of the controllers is caused by not considering the capacity limitation of the transferred-in controllers and ignoring the transmission distance between the switches and the controllers after fault recovery Added technical issues.

The invention discloses a load balancing based performance data acquisition method for a large-scale parallel program. The method is used for performing distributed acquisition on performance data generated by a large-scale parallel program in a high-performance computer and saving the data in an external memory. Nodes in the high-performance computer are divided into three classes of compute nodes, acquisition nodes and control nodes with the method comprising the following steps of: dynamically selecting the needed acquisition nodes with a load balancing algorithm; instrumenting and committing the parallel program on the control nodes through tools; collecting the performance data of the parallel program and sending the performance data to the selected acquisition nodes in a distributedway; receiving and caching the performance data from all the compute nodes in a distributed way and writing the performance data in the external memory. The efficiency and the expandability of the acquisition of the performance data of the parallel program can be improved with the method.