33results about How to "Improve concurrency efficiency" patented technology

The invention discloses a multi-version concurrency control method and device for a database, a server and a storage medium. The method comprises the steps of blocking a transaction number of a current transaction in an exclusive lock mode when the current transaction is started; if a current operation of the current transaction is a write operation, generating a transaction save point; searchingfor a target data page corresponding to the write operation, and locking the target data page in the exclusive lock mode; executing the current operation according to a preset rule. Through this blocking mechanism, the number of locks is not directly related to the number of records modified by transactions, the number of the locks can be greatly decreased, blocking conflicts and blocking resourceconsumption are reduced, and the concurrency efficiency of a database management system is improved.

The invention discloses a multithreading-based web crawler system, which comprises an URL (Uniform Resource Locator) processing module, a web crawling module, a web analysis module and a web storage module, wherein the URL processing module obtains the host name, the port number and the filename of each URL through URL-class processing; the web crawling module carries out partitioning crawling on web contents and stores a captured web into a temporary storage module; the web analysis module extracts the URL, redirects the URL, carries out repetition judgment processing on the URL and deletes repeated the URL; the web storage module judges whether the file is in the presence or not when the file is stored, and the file is directly crawled if the file is not in the presence; if the file is in the presence, contents obtained by crawling the web at the time are more than the contents crawled in the previous time, and the original file is covered; and otherwise, the file is abandoned. The web matched with a regular expression is firstly input, a web request signal is sent, then, a private function is triggered to obtain matched substance, finally, specific information which contains keywords is finally obtained, crawling speed is high, and efficiency is high.

The invention provides a workflow scheduling method and device. The workflow scheduling method comprises the steps of: respectively creating a task queue for each processor, wherein each task queue comprises one or more subqueues with different priorities; inserting a task activity in a workflow example as a task item into the subqueues; extracting a task item from the task queue corresponding to the task item by the processor, wherein the extraction subsequence is realized by extracting the subqueue with high priority; and in the extracted subqueues, extracting the task items at the heads of the subqueue. According to the invention, concurrency efficiency of the workflow system is increased.

The invention discloses a three-stage optimization method oriented to concurrent acquisition of energy consumption data, which is based on a large-scale energy sensing network, and adopts a plurality of energy nodes and an energy management network, wherein the energy management network comprises an energy main controller, a data acquisition terminal and a time processing system; the data acquisition terminal subdivides a data acquisition task DCJ into a plurality of sub-tasks; the time processing system establishes a time Petri network; the energy main controller proposes a three-stage optimization algorithm 3SOA based on a greedy algorithm and a genetic algorithm, so as to solve a dispatching problem of the DCJ; and the 3SOA comprises a first stage optimization, a second stage optimization and a third stage optimization. The three-stage optimization method adopts the 3SOA for solving the dispatching problem of the data acquisition task, obtains the minimal completion time of the DCJ, optimizes a dispatching scheme of the DCJ, can significantly shortens acquisition completion time by adopting the 3SOA, and improves concurrent efficiency; moreover, the application shows that the 3SOA can shorten a data acquisition cycle from 9.8 seconds to 6 seconds and increase the concurrent efficiency by 34.45%.

The invention discloses a method and device for obtaining a spin lock, and the method comprises the steps of carrying out the queuing of a plurality of tasks through an atomic operation function of aCPU; determining a sequence number queue used for representing the sequence of the plurality of task access critical areas, and a queue head sequence number and a queue tail sequence number of the sequence number queue, the queue head sequence number and the queue tail sequence number being used for representing queuing information of spin locks of the critical areas acquired by corresponding tasks; determining an atomic sequence number corresponding to a first task to be accessed to the critical region based on the queue head sequence number; and if the queue tail serial number is equal to the atomic serial number, controlling the first task to obtain the spin lock of the critical region. According to the present invention, the technical problem that an existing spin lock is low in obtaining efficiency is solved.

The invention discloses a data reading and writing method and device. The method comprises the steps that a data writing module finds a target storage unit from a storage queue, the storage queue comprises a plurality of storage units, and each storage unit at least comprises a first field used for storing elements and a second field used for storing read-write identifiers; the data writing modulejudges whether the target storage unit is empty or not according to the second field of the target storage unit; and under the condition that the target storage unit is empty, the data writing modulewrites the element to be written into a first field of the target storage unit and changes a second field of the target storage unit into a first preset value, and the first preset value is used forindicating that the element is written into the target storage unit. According to the method and the device, the technical problem of relatively low concurrence efficiency caused by easy out-of-orderconflict of the blocking queue of the JDK under the condition of high concurrence is solved.

The invention provides a test method and device. The method comprises the following steps: when a system to be tested is tested, determining conflict domains to which test cases needed for testing the system to be tested belong, wherein conflict relations exist between the test cases in the conflict domains, and no conflict relations exist between the test cases belonging to different conflict domains; when the at least two conflict domains are determined, judging whether the at least two conflict domains meet a merging condition or not; when a merging condition is met, merging the test cases in the conflict domain meeting the merging condition to generate a top parent domain; when the combination condition is not met, taking the conflict domain which does not meet the combination condition as a top-layer parent domain; generating a corresponding number of test sub-processes according to the obtained number of the top-layer parent domain; and calling the test case in the top-layer parent domain corresponding to the test sub-process by utilizing each test sub-process so as to test the tested system.

The invention provides a data operating method, a data operating device and a computer device. The data operating method includes the steps: receiving a data operating request; applying for adding a relationship lock for a parent object to be operated, and then applying for adding a structure lock for an object to be operated; performing data operation on the object to be operated according to thedata operating request. According to the method, data concurrent operation can be flexibly controlled through the relationship lock and the structure lock by the aid of a thinner control particle size, and the concurrent efficiency of data operation is improved.

The invention discloses a Redis-based batch job scheduling adaptive method, which comprises the following steps: S1, job loading: creating Redis job channels, receiving to-be-processed batch jobs, and pressing the jobs into the corresponding job channels according to the scale of the jobs; S2, job scheduling: corresponding to a Redis job channel, establishing a response task receiving pool for checking a to-be-processed task of the corresponding channel; S3, job processing: receiving the to-be-processed task of the corresponding channel through the response task receiving pool, and performing job processing; and S4, after the current job is processed, releasing the resources, and returning to S3. According to the method, multiple channels are established by means of the redis cache technology, the concurrence efficiency is improved, and a scheduling mechanism of batch switching connection is realized by means of the mature redis technology; the to-be-processed batches are classified and grouped according to the scale of each batch, and each group is processed in parallel, so that long-time waiting of the to-be-processed batches under extreme conditions is prevented.