62results about How to "Increase the number of concurrency" patented technology

The invention provides a blocked task scheduling method. The method comprises the steps as follows: allocating coroutines for task functions, and calling the task functions in the coroutines; when calling of the task functions is blocked, suspending the coroutines, and setting the coroutines to be in suspended states; monitoring system I / O (input / output) events, finding the coroutines corresponding to the system I / O events, and setting the coroutines to be in running states; finding out the coroutines in the running states, resuming the found coroutines, and correspondingly processing the system I / O events. Besides, the invention further provides a blocked task scheduling device. By the aid of the blocked task scheduling method and device, the concurrency amount of a system can be increased.

The present invention provides an asynchronous concurrent mechanism based pressure test system and test method. The system is provided with a script management apparatus, a test management apparatus, an asynchronous pressure generating apparatus and a resource acquisition apparatus. The script management apparatus selects and edits a test script according to a test object and a test requirement; the test management apparatus selects a pressure test request manner controlled by different pressures according to the test script; the asynchronous pressure apparatus performs a pressure test according to a preset parameter, a set number of generated asynchronous constant rates and a pressure test request of a concurrent mechanism; and finally, the resource acquisition apparatus acquires various data results in real time during the pressure test, so that the test management apparatus can making statistics and analysis on whether the pressure performance of a tested system meets a design index. According to the pressure test system provided by the present invention, an adjustment range of a test parameter of the system is wide, the pressure test request can be generated by selecting multiple test manners and using an automaton concurrent manner in a process, the pressure test data better meeting a real use condition of the tested system can be acquired, and a test result can be verified.

The invention discloses a spatial information cluster cache pre-fetching method for a network spatial information service system. Combined with the long-term epidemic characteristics and short-term epidemic characteristics of tile access, the method comprises the following steps of: establishing a tile transition probability estimation method and establishing a basic Markov model according to theglobal property of tile access and based on the long-term epidemic characteristics of tile access conforming to the Zipf distribution rule; based on the space-time local change of the tile access mode, predicting the tile with large follow-up state access transition probability by use of the basic Markov model, wherein the tile is used as a cache object; and finally uniformly distributing the cache objects into the cluster cache server according to the access transition probability of the pre-fetched tile. In the method, the global property of the tile access mode as well as the space-time local change of the tile access mode are taken into consideration so that the cache pre-fetching object is relatively stable and can adapt to the access distribution change; and thus, the hit rate is relatively high, the cache replacement frequency can be reduced, and the stability and service efficiency of the cluster cache system are improved.

The invention discloses a security and protection cloud platform, and belongs to the technical field of security and protection. The platform comprises an IP camera, a client live streaming source, a handset live streaming source, an open storage cloud service, an on-demand client, an on-demand live streaming server platform, an IDM server cluster, a Web Service interface, a Web business server cluster, a Redis memory database, a MySQL database cluster and an ETS business server. According to the security and protection cloud platform, mature Nginx and Redis frameworks on the Internet are introduced, unification of system data is achieved by using the MySQL cluster technology based on P2Pdirect connection and forward automatic switching technologies, the system adopts a multi-domain mode to allow each user and equipment to be rapidly connected, thus being stronger in expansibility; the number of concurrent user is large, reaching a level of 20W and more per single server, and an Nginx plug-in is used for developing, so that the system is rapid in development and low in cost.

The invention discloses a method for sending / receiving a Radius request, a Radius user terminal and a Radius server, wherein the method for sending the Radius request comprises the steps: the Radius user terminal distributes a user data protocol (UDP) source port and an identifier for the Radius request and sends the Radius request to the Radius server, wherein the Radius request carries the UDP source port and the identifier distributed by the Radius user terminal. In the invention, the UDP source port and the identifier in the Radius request can be jointly served as a combined identifier matching the Radius request, which can solve the problem that the concurrency of the Radius request in relevant technologies can not satisfy the demands when mass users access simultaneously for authentication and billing, thus being capable of realizing the purpose of increasing and sending the Radius request.

The invention provides a resource locking algorithm for a memory read-write process, which is used for read-write operation of a memory resource D with an addressing mark, and the addressing mark comprises an address interval, an authority identifier and a timestamp; wherein the authority identifiers are in three mutually exclusive states of distributable, read-only and write-only; the locking algorithm directs the read operation request to the memory resource of which the permission identifier is read-only; the locking algorithm directs the write operation request to memory resources only written by permission; when the locking algorithm is in write operation, if the write operation aims at the memory space S of the existing content X, the locking algorithm copies the content X in the memory to form a mirror image space L with only write permission, and the mirror image space L is used for replacing the memory space S to respond to the write operation aiming at the content X and replacing the memory space S after the write operation is finished; according to the method, the memory switching algorithm can be used for replacing a mutual exclusion lock to solve the synchronization problem of shared resource mutual exclusion access in the scenes of memory data block single-process (thread) writing and multi-process (thread) reading.

The invention discloses a data processing method and device, a storage medium and an electronic device. The method comprises the steps of receiving a first request, wherein the first request is used for requesting for writing first data into a block chain; storing the first data to a to-be-processed data set; acquiring target data from the to-be-processed data set according to the current data processing capacity of the block chain, wherein the data size of the target data is less than or equal to a data size that the block chain currently allows to process; and writing the target data into the block chain. The technical problem that in related technologies, when the request that requests for writing the data into the block chain is processed, the processing efficiency is relatively low issolved.

The invention discloses a web push method based on an event bus. The method comprises the following steps of S1, service container stack address binding; S2, asynchronous processing of messages; S3, releasing of application service messages; S4, web side message subscription; and S5, event bus pushing. The difficulty of upgrading and capacity expanding can be reduced by the technical scheme, and the workload of upgrading can be reduced so that message sending and receiving are enabled to be more timely; the services can be quickly deployed with increasing of the number of the registered users; the concurrency of the system can be enhanced through asynchronous processing; the TPS (transaction per second) of the system can be obviously enhanced; and service calling is safer.

The invention discloses a memory fragment allocation method and device based on a shared memory. According to the method and device, problem that information is copied for many times in the information transmission and processing process is solved, an information buffer queue is moved from a trusted receiver to the shared memory, and an information queue building and memory fragment allocation method is disclosed on the basis of the shared memory. Copying frequency can be reduced in the information transmission and processing process, system efficiency is improved, and concurrency of a network server is improved; meanwhile, io is separated from logical processing, and lateral expansion capability and flexibility of the network server can be improved.

The invention provides a live broadcast processing method and a live broadcast processing device, applied to a visual network. The method comprises the following steps: receiving, by a visual networking server, a live broadcast releasing request based on a visual networking protocol sent by a virtual terminal, allocating live broadcast channels to the virtual terminal according to the live broadcast releasing request; receiving a live broadcast watching request based on the visual networking protocol sent by the visual networking terminal, sending a live broadcast notification based on the visual networking protocol to the virtual terminal according to the live broadcast watching request; receiving live broadcast video data based on the visual networking protocol sent by the virtual terminal for the live broadcast notification through the live broadcast channels, wherein the live broadcast video data based on the visual networking protocol is obtained by the virtual terminal performingprotocol conversion on the live broadcast video data based on the visual networking protocol; and transmitting the live broadcast video data based on the visual networking protocol to the video networking terminal. The live broadcast processing method and live broadcast processing device can release and watch live broadcasts through the visual network, and the visual network is faster in data transmission speed, smaller in delay, and larger in concurrency, which can enhance the user experience.

The invention discloses a high-concurrency and high-decoupling data processing method and a data middle station system and relates to the field of data middle stations, and the method comprises the following steps that: an Ngnix server receives packaged data and distributes the packaged data to a Publisher instance, the Publisher instance distributes the packaged data to a message queue accordingto the service type; and the Consumer instance acquires the packaged data from the message queue and processes the packaged data according to the service type to obtain a processing result, the Consumer instance stores the processing result to a local database, and the local database provides a local port according to the service dimension and provides a corresponding mapping port in a public network. The method has the advantages that high concurrency, high decoupling and low cost of the data center station system are achieved, the architecture is clear, and expansion is easy.

The invention provides a distributed virtual framework, comprising a server, a CPU (central processing unit), an internal memory, a coprocessor and a client end used for connecting and using a virtual desktop by a terminal user; the server is used for providing the software operating environment and a virtual machine control platform of a virtual machine; the CPU is used for judging a command set processing mode and sending to the coprocessor for computing according to the command set processing mode; and the coprocessor is used for receiving the command sent from the CPU and carrying out corresponding computing. According to the distributed virtual framework, in the whole server, the CPU judges the processing mode of each command set and can issue video processing, network processing and storage processing on the server to the coprocessor for processing according to a coprocessing computing mode and virtual software, thus improving the processing capacity of the CPU and the concurrency number of the virtual machine and reducing the cost.

The invention discloses a data transmission method, a service cluster and a client, which are used for solving the technical problems of relatively low stability and accuracy of file transmission under a high concurrency condition in the prior art. The method comprises the steps that a server receives transmission request description information of a target file from a client and obtains identification information of the target file based on the transmission request description information; the server starts to circularly execute the following operations until the client finishes slicing the target file: inquiring current transmission progress information of the target file based on the identification information, and generating corresponding transmission demand description information; receiving slice data currently sent by the client, wherein the slice data is obtained after the client performs primary slicing processing on the target file based on the transmission demand descriptioninformation; on a local disk array, assembling the assembled part of data and the currently sent slice data, and updating the transmission progress information; and obtaining a target file after allthe slice data are assembled.

The invention relates to a terminal processing method and a terminal processing device. The method comprises the following steps that: a receiving module of a first terminal receives request information which is transmitted by a second terminal and comprises a password and a mode information code; a verification module of the first terminal verifies the password; when the verification passes, the verification module of the first terminal transmits a processing instruction to the relevant module of the first terminal according to the mode information code; and the relevant module of the first terminal processes according to the processing instruction. The terminal processing method and the terminal processing device have the advantages that: a user can process the relevant module of the first terminal according to the request information transmitted by the second terminal, without adding software and hardware modules in the first terminal, so the cost is reduced; and the user can process the relevant module without through a control center, so the concurrent amount and the degree of freedom of the user are improved.

The invention discloses a method and system for building a highly reliable mobile cash register system, which adopts three cash register modes that the normal running line is line one, and two groupsof standby lines are line two and line three respectively, wherein the line one is from a customer, a cash register, a store network, a fourth party payment server to a third party payment server; theline 2 is from the customer, the cashier, the internet of things, the fourth party payment server to the third party payment server; the line 3 is from the customer, the cashier, the 4G network to the third party payment server. Through the design of line 2, the mobile cash register business can be kept uninterrupted under the condition of the outage of the store. Through the design of the line 3, and when the fourth-party server is down, the store will exchange information with the third-party payment server directly, and synchronize the information with the fourth-party payment server afterthe fourth-party payment server is restored. The online rate of mobile cash register service can be approximated to the official online rate of the third-party payment server by this scheme.

This invention relates to a method for making large concurrency flow media to transfer timely. In the method, a server designs the largest number for supporting the flow media, designs the reference task progress, the current error gain and accumulated error gain for them and makes the token potting among the flow media within each frame interval. Every flow medium who gets the token judges if its current window value is zero, if not, it sends and processes data according to the send plan, if it is zero, it computes the error of the general task progress of said flow medium to design a new sending plan and degrades the current window value by 1. Said flow media server can suppress the arrive time, the number server can suppress the arrive time, the number, the wave of the code rate, hardware and network faults so as to increase the concurrency number, stability and usability supported by the server.

The invention provides a big data computing method, a big data computing system, computer equipment and a computer readable storage medium. The big data computing method comprises the steps that to-be-processed tasks are acquired, and a computing job is submitted to a Spark computing engine according to the to-be-processed tasks; the computing job is submitted into a job queue; whether physical resources needed for scheduling the computing job are available in the Spark computing engine is judged; when the physical resources are available in the Spark computing engine, the computing job is scheduled; the computing job is controlled to load computing data, and the computing data is computed to generate a computing result; and the computing result is stored. The method is based on an apportioned computing engine of big data Spark, the advantages of powerful distributed parallel processing and memory computing abilities and quick query of Spark are fully utilized, the parallel quantity ofthe computing tasks is increased, and the efficiency of multidimensional analytical computing is improved.

The invention relates to the field of data transmission, and discloses a depth data transmission method, which comprises the following steps of: encoding depth data into an RGB image according to a predetermined encoding format; transmitting the RGB image by adopting a video stream transmission protocol; receiving the RGB image; and decoding the received RGB image into the depth data according toa predetermined decoding format. The transmission bandwidth pressure can be effectively reduced, the network concurrence number is increased, and the depth information precision loss caused by transmission errors in the transmission process can be effectively reduced.

The invention belongs to the technical field of computers, and particularly relates to a point-to-point network-based data storage method and device, a computer readable storage medium and terminal equipment. The method comprises the steps that a source node divides data to be stored into data fragments, and determines the backup number of the data fragments; the source node determines a first-level storage node corresponding to the sth data fragment, and sends a data storage request to the first-level storage node; the nth-level storage node determines an (n + 1) th-level storage node corresponding to the sth data fragment, and sends a data storage request to the (n + 1) th-level storage node; after the (n + 1) th-level storage node receives the data storage request, the (n + 1) th-levelstorage node downloads and stores the sth data fragment from each target node, adds the node identifier of the (n + 1) th-level storage node into the storage node set, and subtracts one counting unitfrom the counting value until the counting value is 0, thereby reducing the occupation of source node bandwidth resources.

The invention discloses a distributed storage system based on an FTP protocol. The distributed storage system comprises a Windows host FTP Client based on the FTP protocol and provided with a servicesystem; a plurality of Linux extension sets FTP Slave 1, FTP Slave 2... FTP Slave N, used for being responsible for data stream transmission and physical file reading and writing; wherein shared directories of other extension sets are mounted among all the Linux extension sets; and a Linux host FTP Master, used for being responsible for FTP protocol control command processing and load balancing ofall Linux extensions. The invention further provides an implementation method of the distributed storage system. By means of the technical scheme, distributed file reading and writing are achieved, the concurrent retrieval performance is greatly improved, the capacity expansion difficulty and cost are reduced, and the bottleneck problem of file retrieval performance existing in centralized arraystorage is effectively solved.

The invention relates to a container edge proxy-based data message conversion self-description protocol MQTT method, which comprises the following steps of: S1, receiving a plurality of service requests by a middleware layer, and judging a processing sequence of the service requests; s2, performing an authentication process on a protocol message carried by the service request to obtain effective message content; s3, the effective message content is sent to an analysis unit for a message analysis process; and S4, the service request returns a response or reports a state, and state information is updated. The processing pressure of the Internet of Things management platform can be reduced, the communication delay is avoided, and the system response time is shortened.

The invention discloses a method, system and device for executing simulation test tasks and a storage medium, and the method comprises the steps: dividing the simulation test tasks into a plurality ofclusters according to the priority, and sequentially distributing the plurality of simulation test tasks in each cluster to different GPU nodes based on the priority sequence; judging whether the calculation amount of the simulation test task is greater than the calculation force of a single GPU node or not; in response to the fact that the calculation amount of the simulation test task is largerthan the calculation force of the single GPU node, segmenting the rendering part of the simulation test task to form a plurality of subtasks; and allocating the plurality of sub-tasks to the plurality of GPU nodes for execution. Aiming at the problem of high concurrency of simulation test tasks, the tasks are reasonably allocated, so that computing resources can be used to the maximum extent, thetask concurrency quantity is greatly increased, and the completion time of all the tasks is shortened.

The invention is suitable for the technical field of industrial Internet, and provides a processing method and system of an Internet industrial control system and computing equipment, and the method comprises the steps: obtaining an industrial control system recognition task; extracting a fingerprint plug-in and a target IP and a target port of a target industrial control system from the industrial control system identification task; running the fingerprint plug-in, and loading the target IP and the target port to initiate a socket connection request to the target industrial control system; in response to the fact that the socket connection succeeds, sending a specified message to the target industrial control system; receiving a response message returned by the target industrial control system based on the specified message; and acquiring detailed information of the target industrial control system from the response message. The embodiment of the invention can improve the operation efficiency.

The invention discloses a cloud rendering mixed compression transmission system based on a file type. The system comprises a client and a network storage server. The client further comprises a file list, a file compression configuration table, a compression algorithm subprogram based on a file type, a file to be uploaded queue, and a file upload subprogram. The network storage server further comprises file receiving service, a decompression algorithm subprogram based on a file type and storage. According to the system, different types of files, compression ratios and the occupation of CPU computing resources and disk IO influence factors in compression and decompression processes are considered, the related files in rendering tasks are differently treated, and while the improvement of the transmission speed of a file exchanged by the compression transmission of a low compression ratio file is maintained, the resource waste of high compression ratio file compression is avoided.

This invention relates to a method for making large concurrency flow media to transfer timely. In the method, a server designs the largest number for supporting the flow media, designs the reference task progress, the current error gain and accumulated error gain for them and makes the token potting among the flow media within each frame interval. Every flow medium who gets the token judges if its current window value is zero, if not, it sends and processes data according to the send plan, if it is zero, it computes the error of the general task progress of said flow medium to design a new sending plan and degrades the current window value by 1. Said flow media server can suppress the arrive time, the number server can suppress the arrive time, the number, the wave of the code rate, hardware and network faults so as to increase the concurrency number, stability and usability supported by the server.

The invention discloses a web push method based on an event bus. The method comprises the following steps of S1, service container stack address binding; S2, asynchronous processing of messages; S3, releasing of application service messages; S4, web side message subscription; and S5, event bus pushing. The difficulty of upgrading and capacity expanding can be reduced by the technical scheme, and the workload of upgrading can be reduced so that message sending and receiving are enabled to be more timely; the services can be quickly deployed with increasing of the number of the registered users; the concurrency of the system can be enhanced through asynchronous processing; the TPS (transaction per second) of the system can be obviously enhanced; and service calling is safer.

The present invention provides a business sinking processing system and middleware, the system includes a client, a business end and a server, the business end is used to carry the business model and business processing of the client, and is located between the client and the server, It is used to call the data source data to the server, and push the aggregated data to the client according to the client subscription data; the server includes one or more data sources, which are used to provide data to the business end. The invention can reduce the difficulty of development and maintenance, reduce the computing pressure of the client, ensure the stability of the client, and increase the display speed of the client.

The invention relates to the field of data transmission, and discloses a depth data transmission method, comprising the following steps: encoding the depth data into an RGB image according to a predetermined encoding format; transmitting the RGB image by using a video stream transmission protocol; receiving the the RGB image; and decode the received RGB image into the depth data according to a predetermined decoding format. It can effectively reduce the pressure of transmission bandwidth, increase the number of concurrent networks, and effectively reduce the loss of depth information accuracy caused by transmission errors during transmission.

The invention discloses a data processing method and device which are used for improving the operation efficiency of operation objects under the high concurrency condition. The method comprises steps as follows: acquiring a request for operation of an operation object, wherein the request for operation of the operation object comprises a user identification and an operation object identification; inquiring whether user information corresponding to the user identification exists in an external cache, and acquiring the user information corresponding to the user identification from the external cache if user information corresponding to the user identification exists in the external cache; inquiring whether configuration information corresponding to the operation object identification exists in a memory, and acquiring the configuration information corresponding to the operation object identification from the memory if configuration information corresponding to the operation object identification exists in the memory; operating an operation object and generating operation information comprising the user identification and the operation object identification when the user information corresponding to the user identification is matched with the configuration information corresponding to the operation object identification; writing the operation information in a sub-table of a database.