41results about How to "Reduce query pressure" patented technology

The invention discloses a method for querying hot data. The method comprises the following steps that, firstly under a real-time monitoring system, a server end queries the hot data in advance, and the queried data is cached in the internal storage of the server end; when a client queries hot data, the server end can execute data query operation in a pre-cached data; a query result is returned to the client. According to the method, query efficiency and response speed of the client can be improved, and query pressure of a database is reduced.

The invention provides an interactive ElasticSearch depth paging inquiry method, which is characterized by comprising receiving a inquiry request sent by a client; judging whether the inquiry requestis a first inquiry request and is smaller than the paging threshold value, if both of them are negative, carrying out the interactive query mode, aggregating and calculating the data of the exhaustively classified fields other than the query fields, and sending the query criteria to the ElasticSearch server according to the refined field combination selected by the user; encapsulating the resulting data returned by the ElasticSearch server and returning it to the client. The invention can real-timely and dynamically recommend query conditions to interact with the user through the interactive query mechanism, and provides the user with a choice, so that the query is more friendly and efficient.

The embodiments of the invention relate to the technical field of communications, and disclose a resource allocation method and device, and a readable storage medium. The resource allocation method comprises: after determining that the current mode is an attack mode, judging whether the user information stored in a database has user information in a first resource request, the first resource request comprising the user information and identification information of a server requesting access; if having, determining a first resource allocation result according to the user information in the first resource request and the identification information of the server requesting the access, and transmitting the first resource allocation result to a terminal that sends the first resource request; otherwise, skipping allocating resources for the first resource request. The resource allocation device can reject part of resource requests in the attack mode to reduce the query pressure of a query device, thereby reducing the probability that the query device collapses due to excessive query volume, and improving the user experience.

The invention discloses a method and system for searching for online users. A storage unit is set for an SP, and adapted to store the basic information and status information of users. Thus, after receiving an online user search request that carries search conditions from an XDMC, the SP can determine online users meeting the search conditions according to the basic information and status information of users in the storage unit, and send search results that carry the basic information of the online users to the XDMC. The storage unit may be set for the SP or shared by the SP and the XDMS. Because the storage unit set for the SP stores the basic information and status information of users, the SP does not need to interact with the PS to obtain the status information of users and or interact with the XDMS to obtain the basic information of users after receiving an online user search request from the XDMC. This greatly reduces the frequency of communications between communication entities and the communication burden of each communication entity, and increases the response speed of the XDMC.

The invention provides an establishment method, method and system for a time series data combined query cache. Relative to a general key / value cache, the time series data combined query cache established by the establishment method, method and system comprises a storage unit of time series data and a storage unit of query results, wherein the storage unit of the time series data comprises a time data set file and a query time range file, and the storage unit of the query results comprises a unique identification file and a query time range file. The combined query cache has two remarkable improvements regarding to the processing of the time series data: firstly, the hit rate of the cache is remarkably improved, and due to the fact that the combined query cache can identify hit of a part of time intervals, query pressure of a back end storage system is obviously reduced; secondly, storage capacity of a caching system is obviously reduced, and data redundancy in storage is completely eliminated due to the facts that the storage particle size of data in the combined query cache is small and metadata information is rich.

The invention provides a file query method. The file query method comprises the following steps: receiving a file query request of a requester; determining a query priority of each network node storing a file specified by the file query request; selecting network nodes with specified query priorities according to the sequence of the query priorities from high to low; randomly selecting a single group of network nodes with the total number of query nodes from the network nodes with the specified query priority; selecting a group of network nodes from the randomly selected network nodes, and screening out the network nodes meeting the first condition from the selected network nodes to serve as target nodes; and returning the target node to the requester. The invention further provides a filequery device, a computer device and a storage medium. The network nodes between different regional operators in the file query process can be balanced and complementary.

The embodiment of the invention discloses a query method and device, electronic equipment and a computer readable medium. A specific embodiment of the method comprises the steps of obtaining a query request input by a user; determining a first query parameter; determining a target data sequence according to the query request and the first query parameter; and storing the target data sequence and the target attribute information in a cache to obtain the target data sequence in the cache and the target attribute information in the cache. According to the embodiment, the maximum quantity of the target data included in the queried target data sequence is limited by generating the first query parameter, and the target data sequence and the target attribute information are stored in the cache, so repeated query of the same target data sequence is avoided, the query speed is increased, the query processing time is saved for a user, and the query pressure of the server is also relieved.

Embodiments of the invention disclose an on-line terminal management method and device, and an on-line terminal agent. The method comprises the following steps: a first on-line terminal agent receives an attention request message sent by first equipment, and sends an attention agreed response message to the first equipment when an attention is allowed; an attention relationship is established between the first on-line terminal agent and the first equipment; and the first on-line terminal agent synchronizes a state set of the first on-line terminal agent, an action set and information of a state transfer model to the first equipment.

Embodiments of the invention disclose an on-line terminal management method and device, and an on-line terminal agent. The method comprises the following steps: a first on-line terminal agent receives an attention request message sent by first equipment, and sends an attention agreed response message to the first equipment when an attention is allowed; an attention relationship is established between the first on-line terminal agent and the first equipment; and the first on-line terminal agent synchronizes a state set of the first on-line terminal agent, an action set and information of a state transfer model to the first equipment.

The embodiment of the present disclosure relates to the field of data processing technology, and discloses a data processing method, device, electronic equipment, and computer-readable storage medium, wherein the data processing method includes: receiving a query request for data transmission records, and the query request includes target data transmission Status and target query time range; then, according to the target data sending status and target query time range, query in the preset database to obtain at least one data sending record, and the data sending data of each sent data information is stored in the preset database record; then, determine the data information corresponding to each data sending record and the target sending object, and resend the data information corresponding to each data sending record to the corresponding target sending object. The embodiments of the present disclosure effectively improve the accuracy, comprehensiveness and query speed of the query results, and reduce the query pressure on the preset database.

The invention discloses a domain name query processing method, comprising: receiving domain name query request information, obtaining multiple pieces of domain name query request data including a domain name subject and a domain name suffix; grouping the acquired domain name query request data according to the domain name suffix, and The domain names in each domain name suffix group perform asynchronous query in batches based on the preset value; obtain the query interface corresponding to the domain name suffix and the query server address corresponding to the query interface from the configuration file; set the same domain name suffix and use the default value The unit’s domain name query request data is sent to the corresponding query server for domain name query; receive and analyze the domain name query results returned by the query server. The domain name query method of the present invention reduces the number of domain name queries and the query pressure of the query server, speeds up the query response time, and simultaneously improves the success rate of the query.

The invention discloses a method and system for querying demand information based on a distributed database. The method includes acquiring query keywords; acquiring query information in a distributed database according to the query keywords; querying information in the distributed database including keyword-containing content information and keyword-related content information; constructing a multiple hash ring according to the query keyword and query information in the distributed database; the multiple hash ring is used to convert the query The keywords are arranged corresponding to the query information in the distributed database; the to-be-queried keywords input by the user are acquired; and the query information corresponding to the to-be-queried keywords is acquired according to the multivariate hash ring. The invention provides a method and system for querying information based on distributed database requirements, which improves the real-time performance of information querying in the distributed database, and further improves the information query efficiency.

The invention provides vaccine tracing method and system based on alliance chain. The method includes: 1) creation step: respectively creating a blockchain service network and an application layer, wherein the blockchain service network is used for executing a contract method called by signature transaction and performing data synchronization to acquire real block data; 2) submitting step: synchronizing the real block data to the application layer by using the blockchain service network, calling an entity login contract to login an entity account by means of the application layer, and submitting vaccine environment record and vaccine circulation record through the blockchain service network; 3) tracing step: submitting retrieval requirement to the blockchain service network by means of theapplication layer, and according to the retrieval requirement, tracing and acquiring the vaccine environment record and vaccine circulation record by means of the blockchain service network. The method and system can be integrated with a vaccine tracing system which is established in the prior art and satisfies industrial standards, and can effectively acquire the vaccine environment record and vaccine circulation record.

The invention discloses a dynamic displaying method for commodity information, electronic equipment and a storage medium. The method comprises the following steps that: receiving a commodity information displaying request emitted from a client side; according to the commodity information displaying request, inquiring whether a cache stores commodity information which is not out of date or not; andif the cache stores the commodity information which is not out of date, obtaining an appointed quantity of commodity information from the stored commodity information which is not out of date in thecache, and returning the commodity information to the client side for display. By use of the method, the commodity information can be stored by the cache, after the displaying request of the commodityinformation is received, the appointed quantity of commodity information which is not out of date is obtained from the cache, on one hand, the query pressure borne by the server is lowered, and displaying is carried out according to the commodity information stored in the cache in advance, on the other hand, speed for the client side to obtain and display the commodity information is improved, and good user experience is provided for users.

The invention discloses a system and method for reading and writing a database. The system for reading and writing the database of the present invention includes: a writing server, a synchronization unit and at least one reading server; the writing server is used to receive the data written by the user and store it in the writing database; the synchronization unit is used to Synchronize the data written by the user to the reading database of the reading server; the reading database is used to store the reading database for the user to read data. The database reading and writing method of the present invention includes: writing data into a writing database; synchronizing said data to at least one reading database; connecting said at least one reading database to read data. The invention can realize fast reading and writing of the database.

The invention provides an application interaction method and device based on a unified platform, and the method comprises the steps: querying user association information matched with a Token and the effective time of the Token in a platform server from the platform server after an application server receives an interaction request carrying the Token, namely, the effective time of the platform; and then, storing the Token inquired from the platform server and the user associated information in the application server, and setting the effective time of the Token in the application server as the platform effective time of the Token, thereby realizing that the effective time of the Token stored in the application server is consistent with the effective time of the Token in the platform server. In this way, the phenomenon that the pressure of the platform server is large due to the fact that the time of the Token stored in the application server is short is avoided, namely, the query pressure of the platform is reduced, meanwhile, the application server directly verifies the interaction request through the locally stored Token, the platform server does not need to request the Token, and therefore the response speed of the application server is increased.

The invention discloses an incredible cloud center batch resource sharing method based on an exclusive or operation and chaos. In the method, a client A uses an initial secure single sign-on method which is disclosed by the invention and based on a block chain zero-knowledge proof. The method disclosed by the invention is as follows: a central authentication system distributes one globally uniqueidentification for each subsystem; the central authentication system generates one set authorizing accesses to the subsystems for each user; the central authentication system generates one globally unique session token for each user passing login authentication; a user A submits a user name and a password to the central authentication system and applies for the login authentication; the central authentication system verifies the user A; and when the user A accesses a subsystem B, the subsystem B finds a corresponding account in a shared storage area according to a Hash address provided by theuser A and performs verification on the user A according to the following steps. The method disclosed by the invention utilizes the basis of the block chain zero-knowledge proof, the subsystems do notneed to acquire true user information and can judge whether a user has a sign-on right, and the single sign-on security problem in an incredible environment is solved.