376results about How to "Avoid performance bottlenecks" patented technology

The invention discloses a distributed file system metadata management method facing to high-performance calculation. The method comprises the following steps of: 1) establishing a catalogue metadata server cluster, a file metadata server cluster and an object storage server cluster; 2) establishing a global counting-based bloom filter in the catalogue metadata server cluster; 3) when the operation request of a client side arrives, skipping to execute step 4) or 5); 4) enabling the catalogue metadata server cluster to respond to the catalogue operation request of the client side to manage the catalogue metadata; and 5) enabling the file metadata server cluster to respond to the file operation request of the client side to manage the file metadata data. According to the distributed file system metadata management method disclosed by the invention, the metadata transferring problem brought by catalogue renaming can be effectively solved, and the distributed file system metadata management method has the advantages of high storage performance, small maintenance expenditure, high load, no bottleneck, good expansibility and balanced load.

The invention discloses a knowledge graph-based enterprise risk prediction method and system. The method comprises the steps of 1) constructing entities of an enterprise knowledge graph, establishingrelationships between the entities, and extracting knowledge from set data sources to form the enterprise knowledge graph; 2) capturing news information of enterprises and related entities, processingthe captured news information based on the enterprise knowledge graph, marking relationships between the related entities and events, and storing marked information in a structured knowledge base toform an enterprise news information knowledge graph; and 3) performing enterprise risk prediction based on the enterprise knowledge graph and the enterprise news information knowledge graph. The blankof an existing knowledge graph in the field of enterprise risk analysis is filled; and the constructed enterprise knowledge graph and enterprise news information knowledge graph have relatively highpracticality.

This invention publishes a locked distribution method of shared resources in a type of computer integration system including: making more than one node of system members as locking management node of all the shared resources in the system, meanwhile each shared resource matches a locking management node; when one apply asks for applicant or release shored resources lock, and send applicant to the corresponding locking management node of this shored resources, the corresponding locking management node will finish the distribution and release of lock's applicant. This invention also publishes the corresponding structure of computer integration system. The invention can be used for realizing high usable locking service in the integration, and it has a high expansibility and realizes a balance of load.

The invention discloses a creation method of a virtual machine, a management system of the virtual machine and related equipment of the management system. The creation method of the virtual machine, the management system of the virtual machine and related equipment thereof are used for improving performance of the network and an image storage server, and shortening the consumed time to create the virtual machine. The creation method of the virtual machine includes that receiving creation request initiated by a user, confirming a specified type of the virtual machine operating system image requested by the virtual machine creation in an image buffer pool, selecting an objective physical machine to deploy the virtual machine in a hardware resource pool composed of a large number of physical machines according to the creation request of the virtual machine for specified resource requirements, wherein the image buffer pool is built in hardware resources and composed of all physical machines for catching various types of virtual machine operating system images; transferring the specified type of the virtual machine operating system image requested by the virtual machine creation in the image buffer pool to virtual machine running directories of the physical machines; and ordering the physical machine to start the virtual machine.

The invention discloses a load balancing method of virtual machines, related equipment and a trunking system. The network performance of the trunking system can be improved after load balancing. According to certain feasible execution modes, the method includes the steps that a management node determines a plurality of virtual machines in which load balancing is needed in the trunking system; according to the network flow relation among the virtual machines, the virtual machines are integrated into at least one virtual machine cluster, so that the network flow between each virtual machine and at least one other virtual machine in the same virtual machine cluster is larger than or equal to a network flow threshold value, and the network flow between each virtual machine and any other virtual machine in the different virtual machine cluster is smaller than the network flow threshold value; migration suggestions are generated, and the migration suggestions are used for instructing all the virtual machines included in each virtual machine cluster to be migrated or hosted to the same target computational node; the migration suggestions are sent to source computational nodes hosted by one or more virtual machines included in each virtual machine cluster.

The embodiment of the invention discloses a load balancing method in server cluster comprising the following steps: acquiring an address resolution protocol ARP request message sent from a client; if a node server has an ARP response token, returning the client ARP response message, wherein the ARP response token is used for representing the response authority of the node server to the ARP request message; acquiring the load information of each node server in the server cluster, selecting the node server with small load from the server cluster according to the acquired load condition, transmitting the ARP response token to the node server with small load, synchronizing the load information of the node server with the ARP response token to the node server with small load and storing the load information on the load table of the node server with small load. The node server in the embodiment of the invention can flexibly distribute the load according to its own network adapter configuration condition and load condition to improve the server network bandwidth utilization rate, especially to fully utilize all server network adapters and to avoid the performance bottleneck caused by excessive reliance on the load balancing equipment when the server configures multiple network adapters.

The embodiment of the invention relates to a service configuration switching method, device and system. The method comprises the following steps: switching the operating mode into a virtual CM mode based on a virtual mode switching notification message from an OLT (optical line terminal); acquiring the first configuration information based on the DOCSIS (data over cable system interface specification) from an OSS (operation support system); converting the first configuration information into the second configuration information based on the PON (passive optical network) protocol, and performing self parameter configuration based on the second configuration information; and sending the second configuration information to the OLT after configuration. The service configuration switching method, device and system provided by the embodiment of the invention can solve the performance bottleneck of the OLT in the prior art, and improve service dispensing efficiency.

The invention discloses a distributed storage method. According to the distributed storage method, after a first storage node of a distributed storage system receives a content data writing request from an application server, the first storage node determines a second storage node for writing content data according to capacity loads and flow loads of storage nodes in the distributed storage system, notifies the application server of writing the content data into the determined second storage node and creates management data of the content data locally, wherein storage positions, namely storage nodes, of the content data are recorded in the management data. Since content data are distributed on the basis that the capacity conditions and the flow conditions of the storage nodes are taken into full consideration, the problem of resource waste caused by adoption of a unified algorithm for content data distribution is avoided.

The embodiment of the invention provides a method and a device for configuring a server cluster. The method comprises the following steps: running a dynamic routing protocol on a plurality of virtual servers in a virtual server cluster and a router connected with the plurality of virtual servers; sending link information of corresponding virtual IP (internet protocol) from the plurality of virtual servers to the router; generating a corresponding multi-path router for the plurality of virtual servers on the router according to the link information of the corresponding virtual IP; and distributing the traffic flows corresponding to the virtual IP from the router to the corresponding virtual servers through the multi-path router. According to the invention, the problem of performance bottleneck of the single virtual server can be solved, the virtual server cluster can be extended simply and conveniently, and the operation and maintenance cost is greatly reduced.

The invention discloses a metadata management method based on a reconciliation statement, which comprises the following steps: managing the metadata by a metadata server cluster, and after calculating Hash value by using document route, obtaining the corresponding metadata server identification by searching the reconciliation statement. The invention gives out the reconciliation statement regulation strategy under a plurality of typical situations, the novel metadata server cluster structure and introduces the flow of main functional modules. The invention can rapidly eliminate a hot spot metadata server in the metadata server cluster aiming at problems of the metadata management, is convenient for the system adjustment after adding and deleting the metadata server, reduces pay expenses of the time and space required for the metadata management, and improves flexibility of metadata management.

The invention discloses a method and system for implementing a database clustering function. According to the method and system, an existing database system only comprising a single database node is expanded into a database cluster comprising multiple database nodes, different kinds of first-type data are stored in the nodes of the database cluster respectively, on this basis, when a first-type data service request of a sender is received, a matched target database node is assigned for the request based on service elements included in the request, and then the request is distributed to the assigned target database node to be processed. Thus, different kinds of data are stored in different nodes of the database cluster, the database clustering function is achieved through effective service routing, database expansion can be effectively achieved based on the function, and therefore the problem of performance bottleneck brought by a single database is solved.

The invention discloses a linkable distributed network intrusion detection method based on behavior description, which achieves the real-time detection of abnormal flow in a network and timely makes a response by applying methods of the network behavior description and the feature extraction in a distributed detection system and linking with a fire wall, particularly comprising the steps of (a), a detection unit operation process and (b), a master control platform operation process: applying a detection strategy based on the behavior description on distributed detection units to effectively organize the distributed detection units to be an efficient organic whole, thus avoiding the low efficiency and the dispersibility of the existing system scheme, wherein the aim is to solve the low efficiency of the abnormal flow detection on a large network in the practical application and improve the whole detection ability and the protection effect of the detection system for the network.

The invention discloses a method, application and system for managing a server cluster. The method for managing configuration data of the server cluster is applicable to be executed in a management server. The server cluster comprises a plurality of application servers. The method comprises the following steps: receiving first configuration information from a client, wherein the first configuration information comprises an identification of a corresponding configuration item and configuration data of the configuration item; and synchronizing the received first configuration information to a configuration server, so as to facilitate the configuration server to inform one or multiple of the plurality of application servers monitoring the configuration item to acquire the first configuration information.

The embodiment of the invention provides a cache data updating method. The cache data updating method includes: adding an identification of cache data to be updated to a queue for storing information of the cache data to be updated; acquiring the identification of the cache data to be updated from the queue, and querying data from a database according to the acquired identification of the cache data to be updated; using the acquired data to update data, corresponding to the identification of the cache data to be updated, in a cache. The cache data updating method can timely update data in the cache, can reduce the calculation load, and can improve the security of the database. In addition, the embodiment of the invention provides a cache data updating device.

The invention provides a query method based on big data. The method comprises the following steps: storing an image file into an image library of a distributed file system; preprocessing the image file, and extracting an image characteristic; establishing a characteristic index for the extracted characteristic, and storing an index file and the image file into a distributed file database; and when a user submits a query request to a system, a user interface carries out characteristic extraction on submitted living examples or the image characteristic, carries out similarity comparison with the image characteristic in a characteristic index database and then sends a matching result to the user. The invention provides the query method based on the big data, a storage and retrieval process of the image data is optimized through the distributed system, retrieval efficiency is improved, and performance bottlenecks are overcome.

The invention discloses a character string matching method based on a finite state automation and content filtering equipment. The character string matching method comprises the step of combining two sequence-associated states meeting the set conditions in a DFA (Deterministic Finite Automaton) to obtain a combined DFA; the corresponding matching process of character strings comprises the following steps of: sequentially reading characters from a character string database and judging whether the matching is character string matching according to the current state and the read characters; if not, skipping to the next state according to the current state and the read characters; if so, acquiring a matched character string of the current state from the corresponding storage address of the character string, reading next character and judging whether to be matched with next character matched with the character string or not; during matching, continuously reading next character and skipping to the next state until the character strings are successfully matched; and if not, skipping to the next state according to the current state and the read characters. By using the method, the times for accessing a memory during the matching of the character strings can be reduced and the speed and the efficiency for the matching of the character strings can be improved.

The invention discloses a remote automatic upgrading method for digital monitoring equipment. The method comprises the following two parts: the digital monitoring equipment upgrades a client program, and the central server upgrades a service program. The method comprises the following specific steps: after the monitoring equipment is started, transmitting an inquiry request to a central management server at regular intervals; when the port of the central server monitors that the monitoring equipment requires the inquiry request, upgrading the service program and transmitting the information of the corresponding software to the monitoring equipment; after the monitoring equipment receives the information, comparing the information with the locally stored information; and when the comparison result indicates that upgrading is allowable, requiring the central server to download a corresponding file, covering the file and rebooting the system. The invention has the advantages of simple layout and maintenance, high efficiency and flexibility, can use the mode of regular detection by using the digital monitoring equipment, and does not need too complicated manual intervention.

The invention provides a search method based on big data. The method comprises steps as follows: index data of images are stored by the aid of distributed files, all the index data are written into index data files, and then the index data files are distributed and stored in various nodes of a distributed file system; the index data are searched through a MapReduce parallel programming model, a search space is divided into multiple subspaces, then parallel searches are performed on the subspaces through task scheduling, and finally, a final search result is formed through summarizing. With the adoption of the search method based on the big data, processes for storing and searching the data of the images by the aid of the distributed system are optimized, the search efficiency is improved, and the performance bottleneck is overcome.

The invention discloses a multi-unmanned-aerial-vehicle formation consistency control method based on event-triggered communication. With the distributed control architecture, the control center in the centralized control architecture is removed, so that no phenomenon of formation failure due to the destruction of the central control node exists. Computing tasks of the UAV formation are controlled, the tasks are distributed from a control center to onboard computers of the member UAVs, thereby greatly improving the overall computing power of the formation and breaking through the performance bottleneck caused by the computing power; with introduction of an event trigger communication mechanism, the algorithm does not need to acquire cooperative state information in each control cycle continuous communication, and only intermittent communication needs to be carried out under a trigger communication condition. Compared with a consistency algorithm based on continuous communication, the algorithm has the advantages that the network information transmission quantity can be greatly reduced, the limited communication bandwidth is used for transmitting more task information, and the dependence on the performance of a communication system can be reduced The algorithm has certain tolerance to transient communication interruption or signal interference, and engineering implementation isfacilitated.