106 results about "Single server" patented technology

The invention provides a server energy management method and device based on a BMC (Baseboard Management Controller). The method comprises the following steps: collecting hardware state data of a server in real time through the BMC; preprocessing the hardware state data, removing abnormal values and supplementing missing data to determine standardized hardware data; inputting the standardized hardware data into a load prediction model to determine a load prediction result; generating an energy efficiency evaluation report according to the standardized hardware data and the load prediction result; determining a dynamic energy adjustment strategy based on the energy efficiency evaluation report; and executing the dynamic energy adjustment strategy, and dynamically adjusting hardware operation parameters of the server so as to realize optimization of energy consumption. According to the method, the hardware control capability and the intelligent analysis algorithm of the BMC can be deeply fused, refined, automatic and intelligent management of server energy is realized, the energy utilization efficiency of a single server is remarkably improved, and conversion from extensive power supply to refined energy consumption management is realized.

The invention discloses a server case unit and a server cabinet, and relates to the technical field of servers, the server case unit comprises a computing node and a switching node, the computing node is orthogonally connected with the switching node, the switching node is provided with a data interface, the data interface is used for being connected with another data interface, and the computing node is connected with the switching node. The computing node is used for realizing networking among a plurality of server case units, the computing node comprises an external interface group, and at least one external interface in the external interface group is used for being connected with a server machine head, so that the ratio of the computing node to the server machine head is adjusted. Through the design of the modularized server case unit, a single server case unit can be used as an independent high-density reasoning server; and convenient networking can be performed in the server cabinet through the data interface on the back surface of the server case unit, so that a large-scale training cluster can be constructed. And an external interface group is designed in the computing node, so that the flexible configuration of the ratio between the server head and the computing node is realized.

The invention provides a bullet screen live broadcast interaction method, a server, a program product and a storage medium, and relates to the technical field of real-time communication and online live broadcast. High-concurrency bottlenecks which are prone to occurring under an original centralized service architecture are processed in a decentralized mode, a single server and a synchronous service cluster are not relied on any more, the system is only responsible for data synchronization and state synchronization between a client side and a server side, long connection communication between the client side and the server side is maintained, and specific service logic is not executed; and the business service cluster is responsible for processing all back-end business logics and decoupling connection maintenance and business processing, so that the problem of long connection interruption of all users caused by business service restarting or faults is avoided. The service stability and high availability of the system in a multi-user live broadcast scene are improved, the user does not perceive abnormal conditions in the live broadcast process, and the user experience is improved.

This invention provides a dynamic update method for single-server preprocessing of Private Information Retrieval (PIR), including an offline preprocessing stage, a dynamic database update stage, an online query stage, and an update stage. Based on the existing single-server preprocessing PIR retrieval scheme Piano, this invention adds support for dynamic database operations, supporting CRUD (Create, Read, Update) operations on the database. Unlike the cumbersome preprocessing work of comparable schemes, this invention introduces incremental preprocessing technology, allowing the original preprocessing to continue to be used after database changes, while incurring update costs proportional to the number of changes, rather than proportional to the size of the database. Compared to existing technologies, this invention reduces total communication costs, client computation time, and client storage space.

The invention provides a multi-keyword private information retrieval method and system based on inverted indexes, and belongs to the field of information security and privacy computing. The method comprises the following steps that: a server splits an original database into a reverse index part of a keyword-index set and a key value pair part of an index-data value; mapping the index set into an integer by utilizing Godel coding; processing the data by adopting probability batch coding and binary random linear coding and publishing a hash function; the client maps a multi-keyword query into a bucket by using a hash function to generate a query vector, and the query vector is sent to the server after being subjected to fully homomorphic encryption; the server performs homomorphic calculation on a matching result in a ciphertext state and returns the result; and the client obtains a matching index set after decryption verification, and then initiates a second round of query to obtain final data. According to the method, symmetric privacy protection of multi-keyword non-primary key query in a single-server environment is realized, leakage of query content and database information is effectively prevented, and private information retrieval security is improved.

The invention discloses a dynamic single-server client preprocessing private information retrieval method, and belongs to the technical field of private information retrieval. The method comprises a preprocessing stage: a client generates a private prompt and stores the private prompt in a local device; a query and response stage: the client accelerates query by using a private prompt; in the refreshing stage, the client repositions the used private prompt and updates the private prompt stored locally; and a dynamic updating stage: when the server updates or deletes the original data item or adds a new data item, the client directly updates the private prompt stored locally. According to the method, the client storage time and the server online calculation time are optimally balanced, and meanwhile, efficient updating of the dynamic database is supported. And when the database of the server dynamically changes, the private prompt stored in the local equipment of the client can be updated within the expected PRP calling time, so that the private prompt updating efficiency in a dynamic database scene is remarkably improved.

The application discloses a socket.io-based message pushing method and device and related components. The message pushing method comprises deploying a load balancer and multiple socket.io server nodes, establishing a connection between a terminal device and any of the socket.io server nodes through the load balancer, integrating a Kafka message queue into each socket.io server node, and using the Kafka message queue execution platform to perform a message pushing process to the terminal device. The application uses the load balancer to distribute the connection requests of a large number of terminal devices to different socket.io server nodes, disperses the connection pressure of a single server, and improves the connection stability of each socket.io server node and the terminal device. Moreover, the integration of the Kafka message queue can timely locate the socket.io server node connected to the target terminal device, and then accurately push the message to the target terminal device through the socket.io server node, thereby improving the pushing efficiency and accuracy in the scenario of a large number of devices.

The invention discloses a method for starting up a field device (FG1, FG2), comprising the following steps carried out by the manufacturer of the field device (FG1, FG2): operating a ticket server (TS); and linking the field device (FG1, FG2) to the ticket server (TS), the field device (FG1, FG2) having no account; and comprising the following steps carried out by the operator: creating an account having access data and having specific access rights, in particular having a user name and a password, and assigning the account to the field device (FG1, FG2) by means of the ticket server (TS); generating one or more tickets (T) by means of the ticket server (TS) for the new account; transporting the ticket (T) to the field device (FG1, FG2); checking the ticket (T), by means of the field device (FG1, FG2), as to whether the ticket (T) was actually created by the ticket server (TS); storing the account on the field device (FG1, FG2) if the check is positive; and granting access to the field device (FG1, FG2) in accordance with the access rights after successful login with the access data on the field device (FG1, FG2) by the user. The invention also discloses a system for carrying out the method.

The invention provides a risk list supervision system and method, the risk list supervision system comprises a database, a risk analysis unit, a risk list generation unit and a server, the database is used for storing account data of accounts and transaction cash flow data; a transaction risk model of the risk analysis unit is used for analyzing transaction cash flow data to generate at least one risk factor corresponding to the transaction cash flow data and calculating the weight of the at least one risk factor to generate a risk score, and when the risk score is larger than a risk standard threshold value, the risk list generation unit generates and sends a risk account list, and the risk account list is used for generating a risk account. The server marks the account corresponding to the account data in the risk account list as a temporary forbidden state; therefore, the risk list supervision system provided by the invention can identify the high-risk account timely, efficiently, quickly and automatically, so as to improve the efficiency and reduce the labor cost.

The application provides a display device and display system capable of displaying information of each node in a server, comprising: a display terminal, which is installed in a monitoring cabinet and is provided with a plurality of USB interfaces at the rear end of the cabinet; the display terminal is in communication connection with each USB interface through a cable, and each USB interface is connected with a server node; each server node is configured with an IP address of an I2C protocol; the display terminal is connected with the server node and communicates with a BMC module of the server node through the I2C protocol. The BMC module of each server node is an information acquisition party of a node state, and the BMC module sends the node state information to the display terminal; the display terminal is an information receiving party of each node state, receives the node state information sent by the BMC module, and displays the information in real time. Through the application, a computer room operation and maintenance personnel can check the information of a single server node in the cabinet without going back to the console to check.

The invention relates to the technical field of hydraulic engineering monitoring, in particular to a hydraulic engineering operation state intelligent diagnosis system based on big data. A hydraulic engineering operation state intelligent diagnosis system based on big data comprises a multi-source data perception and fusion module which constructs a real-time data acquisition network covering basin hydraulic engineering total factors through a plurality of engineering detection devices and an environment perception terminal, acquires engineering environment data, and sends the engineering environment data to a cloud server; engineering environment data are processed through edge computing nodes, and then multi-source heterogeneous data are subjected to fusion governance by using a Hadoop / Spark big data platform; a diagnosis scoring module, a toughness evaluation module, an edge-cloud collaborative computing module and an intelligent decision early warning module. According to the method, a parallel special server is arranged through an innovative hardware architecture, multi-dimensional specialized synchronous calculation is realized, the processing bottleneck of a single server is effectively broken through, and a stable physical basis is constructed for cluster collaborative diagnosis.

A method for improving the production efficiency of a digital printing linkage line comprises the following steps that 1, in a client side, the carton size and a pdf printing file are input, and an order is submitted; 2, the server receives the order and gives a URL (Uniform Resource Locator) of the pdf; 3, printing a production work order, wherein a two-dimensional code on the work order comprises order information; 4, the digital printing linkage line is composed of a digital printer, a front edge paper feeder and a grooving machine; the digital computer is interconnected with the PLC, the leading edge PLC and the grooving machine PLC through a local area network, 5, the digital computer scans a code work order and sends order information to each PLC module on the local area network; each PLC module receives the order information and automatically calls the order; the digital computer automatically downloads the pdf and carries out sharpening processing; and 6, production can be carried out after order adjustment of each module is finished. Paper feeding is started, and paper feeding, printing and grooving procedures of paper sheets are completed at a time through the linkage line; a worker only needs to scan the code once during carton production, parameters do not need to be input manually, and the production speed is increased.

To enable issuers to offer a variety of services without burdening merchants or acquirers. [Solution] The credit management method of the present invention includes the steps of: when a commercial transaction occurs, the issuer server receives transaction information from an acquirer server which contains merchant identification information that identifies a merchant where the commercial transaction took place, the credit card number used in the commercial transaction, and acquirer identification information that identifies an acquirer server which received the transaction information from the merchant terminal; and the issuer server notifies the acquirer server of the credit result for the received transaction information and stores the commercial transaction separately from other commercial transactions if the combination defined by the credit card number, the acquirer identification information, and the merchant identification information matches predetermined conditions.

This invention discloses a cryptographic generation system, method, and apparatus based on distributed quasi-prime numbers. A second server randomly determines a second random integer within a second numerical range that satisfies a preset second extraction condition. When each third modulo result and each corresponding fourth modulo result of the second random integer satisfies a preset second synthesis condition, a quasi-prime number is determined based on the second random integer and a first random integer determined by the first server. The first random integer is an integer randomly determined by the first server within a first numerical range that satisfies the preset first extraction condition, and each corresponding first modulo result and each corresponding second modulo result satisfies the preset first synthesis condition. The first and second servers only need to communicate once to determine the quasi-prime number, reducing communication resources and the resources required for a single server to determine the quasi-prime number, improving the efficiency of prime number generation, and thus improving cryptographic generation efficiency.

This invention relates to a multi-point concurrent UDP bandwidth measurement method, apparatus, and storage medium, applied in the field of network communication technology. It includes: concurrently connecting a client to multiple speed test servers and aggregating and calculating test data from all paths, effectively avoiding interference from momentary congestion of a single network path or load bottlenecks of a single server on the measurement results, enabling the measurement results to more realistically and stably reflect the client's local network access limit capacity; employing a two-stage rate control algorithm combining "target verification and binary search," first attempting to verify the user's expected target bandwidth. If network conditions are good, the test can be completed in one go, greatly improving efficiency; if the target is not achieved, binary search is used to quickly approach and converge to the limit bandwidth value within a fixed number of rounds. This automated detection process significantly improves testing efficiency and automation.

The present application relates to the technical field of water cooling heat dissipation, in particular to a server water cooling heat dissipation online control method, system and device. The method analyzes the response correlation between the flow data and load monitoring in the cooling branch of a single server node, combines the change trend between the cooling liquid flow data and the temperature data of the server node, and evaluates the thermal response matching degree. According to the relationship between the thermal response matching degree of the server nodes and the temperature difference change of the inlet and outlet pipes, the heat transfer balance deviation is analyzed. According to the synchronous response of the loop pressure data change in each server node branch, the control deviation influence coefficient is obtained, and the cooling liquid distribution is regulated and controlled comprehensively. The present application adjusts the cooling liquid distribution control of different nodes by periodically monitoring the heat dissipation balance and execution response between nodes, improves the reasonable balance of the cooling liquid distribution in the water cooling heat dissipation process, and reduces the system energy consumption.

The invention discloses a message pushing method and device based on Socket.io and a related component. The message pushing method comprises the steps that a load balancer and a plurality of Socket.io server nodes are deployed, and connection between terminal equipment and any one of the Socket.io server nodes is established through the load balancer; and integrating a Kafka message queue into each socket. Io server node, and executing a message pushing process of a platform to the terminal equipment by utilizing the Kafka message queue. According to the method, the load balancer is utilized to distribute the connection requests of the massive terminal equipment to different socket.io server nodes, so that the connection pressure of a single server is dispersed, and the connection stability of each socket.io server node and the terminal equipment is improved. And the Socket.io server node connected with the target terminal equipment can be positioned in time by integrating the Kafka message queue, so that the message is accurately pushed to the target terminal equipment through the Socket.io server node, and the pushing efficiency and accuracy in a massive equipment scene are improved.

The application provides an implementation method for limiting the bandwidth of IPv6, wherein the address of the IPV6 is placed on a CVK virtual machine or a single server in a VPC network; when the IPV6 accesses an external network, the data packet flow path is Leaf switch, Spine switch, VPC border firewall, Internet interface and router in sequence; the router and the VPC border firewall limit the bandwidth and flow of the IPV6; when the IPV6 is sold, the shared bandwidth is added; and when the IPV6 stops serving, the shared bandwidth is removed. The application has the beneficial effect that the router and the VPC border firewall are used to limit the bandwidth and flow of the IPV6, and the QOS strategy of the uplink and downlink ports on the router is used to realize the purpose of the flow limitation and flow statistics selling of the IPV6.

The invention discloses a multi-dimensional performance evaluation method and system for a computing power server, and relates to the field of server performance data analysis. According to the multi-dimensional performance evaluation method and system for the computing power server, existing computing power server resources are constructed into a single server system, a server cluster system and a server distributed system; selecting a server system most suitable for running the task according to the characteristics of the test task, and selecting an optimal value from four dimension indexes of deployment time consumption, running time consumption, upgrading time consumption and unloading time consumption to evaluate the performance of the existing computing power server; comprehensiveness of performance testing and full utilization of existing resources are achieved, and the final evaluation result can reflect the optimal performance state which can be achieved under the existing computing power server resource condition; and the accuracy of the performance test is improved.

The invention relates to the technical field of water-cooling heat dissipation, in particular to a server water-cooling heat dissipation online control method, system and device. According to the method, the thermal response matching degree is evaluated by analyzing response association between flow data and load monitoring in a cooling branch of a single server node and combining the change trend between cooling liquid flow data in a current node and temperature data of the server node; according to the thermal response matching degree between the server nodes and the relation between the temperature difference changes of the inlet and outlet pipelines, the heat transfer balance deviation is analyzed; according to the synchronous response condition of loop pressure data change in each server node branch, a control deviation influence coefficient is obtained, and cooling liquid distribution is regulated and controlled by integrating heat transfer balance deviation. The cooling liquid distribution control of different nodes is adjusted by periodically monitoring the heat dissipation balance condition and the execution response condition between the nodes, the reasonable balance of cooling liquid distribution in the water-cooling heat dissipation process is improved, and the system energy consumption is reduced.

The invention discloses an Internet of Things equipment simulation online test method, system and equipment and a medium. The method comprises the steps that information of equipment to be tested is collected and then imported into a database through a data information issuing platform; an equipment simulation module pulls written equipment information from a database to create an event cycle based on syncio, an independent simulator is instantiated for each piece of Internet of Things equipment, an exclusive asynchronous coroutine is created for each simulator, each simulator creates an independent MQTT client side through aiomqtt to monitor a message channel of an instruction issued by an Internet of Things platform, and the message channel is sent to the Internet of Things platform; and analyzing and processing the arrangement task, constructing equipment attributes and event data, and reporting the equipment attributes and event data to the Internet of Things platform. According to the method, a simulator is adopted to replace real equipment for testing, multi-coroutine concurrent scheduling in a single thread is realized through an asycio asynchronous programming architecture and in combination with an aiomqtt asynchronous MQTT client, the thread switching overhead of a traditional synchronous architecture is avoided, a single server can support parallel simulation of more equipment, and the method is simple and convenient to operate. And the online test requirements of massive equipment can be met while the test efficiency is improved.

This invention discloses a multi-dimensional performance evaluation method and system for computing servers, relating to the field of server performance data analysis. By constructing existing computing server resources into single server systems, server cluster systems, and distributed server systems, this invention selects the most suitable server system for running the test task based on its characteristics during performance testing. It then evaluates the performance of existing computing servers by selecting the optimal value from four dimensions: deployment time, running time, upgrade time, and uninstallation time. This achieves comprehensive performance testing and full utilization of existing resources, ensuring that the final evaluation results reflect the best performance achievable under existing computing server resource conditions, thus improving the accuracy of performance testing.

The invention relates to a multipoint concurrent UDP (User Datagram Protocol) bandwidth measurement method and device and a storage medium, which are applied to the technical field of network communication, and comprise the following steps of: concurrently connecting a plurality of speed measurement servers through a client, and carrying out aggregation calculation on test data of all paths, the interference of instantaneous congestion of a single network path or the load bottleneck of a single server on the measurement result is effectively avoided, so that the measurement result can reflect the ultimate bearing capacity of the local access network of the client more truly and more stably; a two-stage rate control algorithm combining'target verification and binary search 'is adopted, firstly, a target bandwidth expected by a user is tried to be verified, if the network condition is good, the test can be completed at a time, and the efficiency is greatly improved; and if the target is not achieved, the target is quickly approached and converged to the limit bandwidth value in a fixed and few rounds through binary search, and the automatic detection process remarkably improves the test efficiency and the automation degree.

The present application relates to the technical field of information security, and especially relates to an index-based single-server anonymous query method, which is applied to an index-based single-server anonymous query system, the system comprising a database server and at least one user terminal in communication connection with the database server, and the method comprising the following steps: S1, an inadvertent transmission protocol between a pre-constructed user terminal and a database server receives a first input query index of a user and returns a decryption parameter corresponding to the query index to the user terminal; S2, the user terminal acquires a query result corresponding to the query index according to the decryption parameter and a grouping data packet sent by the database server to the user terminal; the grouping data packet is obtained by the database server from n stored data by using the Chinese remainder theorem for preprocessing; each stored data comprises data information and a query index corresponding to the data information.

The embodiment of the invention discloses a data distribution method and device, a server and a computer readable storage medium, and the method comprises the steps: a local selective forwarding unit server in a distributed selective forwarding unit server cluster receives target distribution data, and determines regional information of a subscription client corresponding to the target distribution data, determining a target selective forwarding unit server from a distributed selective forwarding unit server cluster based on the regional information, and when determining that the target selective forwarding unit server is a remote selective forwarding unit server, sending the target distribution data to the remote selective forwarding unit server, and sending the target distribution data to the subscription client by the remote selective forwarding unit server in the distributed selective forwarding unit server cluster. Therefore, the technical problem that when a single SFU server accesses clients in different regions, due to remote communication, the real-time data communication quality of the clients is low is solved.

Methods and server systems for improving accuracy of authorization optimizer are described herein. Method performed by server system includes receiving a Non-Sufficient Funds (NSF) error message from an acquirer server. Method includes accessing historical transaction data from a transaction database. The historical transaction data includes transaction related information associated with a plurality of users. Method includes generating a plurality of transaction features associated with the user based on the historical transaction data. Method includes determining via an authorization optimizer model, an optimal time slot from a plurality of time slots for the user based on the plurality of transaction features associated with the user. The optimal time slot indicates an optimal time window for the acquirer server to transmit an upcoming recurring payment request to the payment account of the user. Method includes facilitating transmission of a notification message including the optimal time slot to the acquirer server.

The application provides a method, a server, a program product and a storage medium for bullet screen live broadcast interaction, and relates to the technical field of real-time communication and online live broadcast. High concurrency bottlenecks that are prone to occur under a centralized service architecture are dispersedly processed, and a single server is no longer relied on. A synchronization service cluster is synchronized, which is only responsible for data synchronization and state synchronization between a client and a server, maintains long connection communication between the client and the server, but does not execute specific business logic. A business service cluster is responsible for processing all back-end business logic, decouples connection maintenance and business processing, and avoids the problem that all user long connections are interrupted due to business service restart or failure. The system service stability and high availability under a multi-person live broadcast scene are improved, and users have no perception of abnormal conditions during the live broadcast process, and the user experience is improved.

The invention relates to the technical field of distributed optimization algorithms with high communication efficiency, in particular to a server cluster-oriented distributed optimization solution method with high communication efficiency, which comprises the following steps of: S1, planning a data set M to be processed, and quantitatively distributing the data set M to a cluster consisting of N servers for processing, establishing a global cost function model according to the cost function model of the single server, and establishing a final objective function model to be optimized and solved; s2, designing a novel distributed optimization solution algorithm based on an event triggering mechanism, and proposing a novel distributed event triggering condition; s3, specifically explaining an execution process of an event triggering mechanism in a distributed optimization solution algorithm; s4, determining and verifying algorithm parameters; with the adoption of the method, the dependence on cooperative computing between the neighbor real-time state and the nodes is avoided in the execution process; the event trigger control strategy can greatly improve the overall execution efficiency of the algorithm on the premise of not influencing the convergence speed and convergence accuracy of the algorithm.

This application discloses a stress testing method, apparatus, device, and computer-readable storage medium, belonging to the field of system performance testing technology. The method includes: determining the concurrency level based on the number of centralized test data sets of the system under test and the number of concurrent processing data sets of a single server in a server cluster; generating a stress testing application corresponding to the server cluster; sending the stress testing application and the concurrency level to the server cluster; and obtaining the stress testing data fed back by the server cluster. The above technical solution can easily obtain the stress testing data fed back by the server cluster and does not require the use of a dedicated stress testing machine, thus simplifying the stress testing method.

The invention discloses a business data distributed calculation implementation method and device, equipment and a storage medium. The method comprises the following steps: preprocessing a real-time business data set to obtain a plurality of preprocessed data streams; each computing node in the factor computing cluster subscribes to obtain the preprocessed data stream from the computing node, performs operation to obtain first-stage factor data and sends the first-stage factor data to a factor storage node in the factor storage cluster; each factor storage node processes the first-stage factor data to obtain corresponding second-stage factor data; and a signal calculation node in the signal calculation cluster obtains two-stage factor data from the factor storage node, and inputs the two-stage factor data into the multi-modal model to obtain a signal processing result. According to the embodiment of the invention, a multi-level multi-node architecture is realized through the factor calculation cluster, the factor storage cluster and the signal calculation cluster in the server cluster, and a single server is prevented from being used for data processing, so that the single-node calculation pressure is effectively relieved, and the signal generation accuracy and the strategy execution stability are improved.