274results about How to "Reduce response delay" patented technology

The invention provides a method for dispatching I/O of an asymmetry virtual machine based on multi-core dynamic partitioning, which comprises the following steps: when a system is started, firstly an I/O dispatching initialization module completes the configuration of dispatching parameters and the allocation of a buffer zone; subsequently, when the system runs, an I/O status monitoring and information processing module dynamically collects and processes the runtime parameters of driving Domain and customer Domain, carries out statistics on parameters such as types, time and frequency of happened I/O event, and then sends the statistics result to a dispatching decision module for judgment; and the dispatching decision module makes dispatching decision according to preset conditions, and gives out commands to a multi-core dynamic partitioning module to execute the operation of multi-core dynamic partitioning or recovering, thus changing the core partitioning mode and dispatching status of a processor of the system and leading the using mode of processor resources and the dispatching switching strategy of Domain to better adapt to I/O load requirements of the current system, and achieving the purpose of optimizing I/O performance.

Various concepts which further improve multi-view/layer coding concepts, are described.

A display device includes: a display unit including pixels coupled to scan lines for transmitting scan signals, data lines for transmitting data signals, and light emission control lines for transmitting light emission control signals; a scan driver; a data driver; and a light emission driver. Each pixel includes: an OLED; a driving transistor to transmit a driving current corresponding to a data signal to the OLED; a first transistor to transmit the data signal to the driving transistor according to a first scan signal; a second transistor to apply a first power source voltage to a first electrode of the driving transistor according to a second scan signal, during an initialization period for initializing a gate electrode voltage of the driving transistor; and a capacitor including a first electrode coupled to a gate electrode of the driving transistor and a second electrode coupled to a first power source supply.

The invention discloses a multi-unmanned aerial vehicle auxiliary edge computing resource allocation method based on task prediction. The method comprises the following steps of: firstly, modeling a communication model, a computing model and an energy loss model in an unmanned aerial vehicle auxiliary edge computing unloading scene; modeling a system total energy consumption minimization problem of the unmanned aerial vehicle auxiliary edge computing unloading network into task predictable process of terminal devices; obtaining prediction model parameters of different terminal devices by adopting centralized training through accessing historical data of the terminal devices; obtaining a prediction task set of the next time slot by utilizing the prediction model based on the task information of the current access terminal devices; and based on the prediction task set, decomposing an original problem into an unmanned aerial vehicle deployment problem and a task scheduling problem for joint optimization. The response time delay and completion time delay of the task can be effectively reduced through the deep learning algorithm, so that the calculation energy consumption is reduced; anevolutionary algorithm is introduced to solve the problem of joint unmanned aerial vehicle deployment and task scheduling optimization, the hovering energy consumption of the unmanned aerial vehicleis greatly reduced, and the utilization rate of computing resources is increased.

The invention discloses an RDMA (Remote Direct Memory Access)-based distributed memory file system. In an initialization phase of the distributed memory file system, memory used by a cluster for filestorage is uniformly divided, and is registered to network cards to support direct memory access of remote-end nodes, and then a distributed shared memory pool is constructed; on the distributed memory sharing pool, file indexing and file data block indexing are respectively carried out through two-level hash indexing to provide query services for the file system; and a request of a client is processed through a self-identifying remote-procedure-call (RPC) method, and a processing result is returned. The system has the following advantages: reducing data copying of files in reading/writing, reducing response delay, and improving overall efficiency of file access in program software.

The invention discloses a quick voice interaction method and system. The method includes receiving voice input of a user in real time; performing real time voice identification on the voice input information and acquiring a voice identification result; performing real time semantic comprehension on the current voice identification result based on a pre-constructed semantic comprehension model andacquiring a model output result; determining whether a correct and complete semantic comprehension result is in presence or not according to the model output result; generating an interaction result according to the semantic comprehension result and sending the interaction result to the user as a feedback if the correct and complete semantic comprehension result is in presence; continuing real time voice identification on the voice input information if the correct and complete semantic comprehension result is not in presence. According to the invention, feedback delay of voice interaction canbe reduced and user experience can be improved.

Response latencies in a system are reduced by performing a plurality of auctions in parallel using respective processing threads. An ad call associated with an impression for a webpage requested by a user computer is received by the system. Impression information is obtained from the ad call, which is then used to determine a subset of multiple potential impression providers that are to participate in an auction for the impression. Behavior models are used to determine the subset, while floor or reserve price models are used to set a floor/reserve price for each potential impression provider in the subset. Bid requests are sent out in parallel to the subset of potential impression providers using the respective floor/reserve prices, with a thread allocated to each bid request. The models can be further trained on information obtained from the received bid responses and impression information.

The invention discloses a multi-controller dynamic deployment method of a software defined spatial information network. In the case of flow surge of switching nodes, the overload switch node can be migrated to an adjacent non-overload area by means of the collaboration of the controllers to realize the dynamic migration of the switch so as to reduce the processing pressure of the controllers and realize load balance of multiple controllers. A multi-target optimization method based on switch flow r, an interaction delay t and a controller volume f is provided in view of the migration problem of the overload switch by the controllers, a proper migration switch is selected according to the three parameters, and a target controller is selected according to the supportable residual load of the target controller; and with respect to the underload controller, the target is migrated by considering the requirement that the load of the other controllers can bear the migration of all switches in the underload controller, and then the underload controller is closed. By adoption of the multi-controller dynamic deployment method disclosed by the invention, the network response delay is shortened, the network throughput is improved, and the load balance among the controllers is guaranteed.

The invention discloses a mapping table management method for a solid state disk and the solid state disk. The method is characterized in that an SCM is added into the solid state disk, a hard disk controller regards the SCM as a solid state disk data cache, and a mapping table used for recording a mapping relation between a specific physical address of an NAND and a logic address accessed by a host is created and stored in the SCM; and when the host writes host data into the solid state disk, the data is written into the SCM first to be cached, and the mapping table is updated synchronously after the data cached in the SCM is actually updated to the NAND or the mapping table is updated synchronously when the data is written into the SCM to be cached. By storing the mapping table in the SCM, the mapping table no longer needs to be stored in the NAND, occupation of space of the NAND is avoided, and therefore the problem of writing amplification caused by frequent updating of the mapping table is avoided. Meanwhile, the operation of loading an initial mapping table from the NAND after startup is omitted, and startup response delay is greatly shortened.

The invention discloses a cache sharing method for a content delivery network, which comprises that: after receiving a service request, a first cache in a cluster determines a content block with contents corresponding to the service request if not being hit; the first cache determines a second cache for processing the content block according to correspondence between the content blocks and caches in the cluster; and the first cache transmits a query message to the second cache to obtain the contents. The embodiment of the invention also provides a cache sharing device. Compared with an Internet cache protocol (ICP)-based cache sharing way, the way of only transmitting the query message to the cache for processing the same content block shortens response delay, and increases the service request processing speed. In addition, the query message is judged to be transmitted to which cache by searching for the correspondence between the content blocks and the caches, and a memory space occupied by the correspondence between the content blocks and the caches is much smaller than that occupied by abstracts; therefore, compared with an abstract protocol-based cache sharing way, the embodiment of the invention reduces memory overhead and searching overhead.

A method for path predictive congestion avoidance includes receiving activity data for a plurality of cells, determining dominant paths for each of a plurality of cells based on the activity data, predicting a future network condition based on the dominant paths, and allocating network resources based on the predicted future network condition.

The invention discloses a quantum information feature extraction method based on CTP financial data. Data issued by an exchange is processed by means of a quantum information algorithm model, a superposed-state quantum information presentation mode is firstly adopted to effectively describe the diversity of transaction data, description and processing by means of a quantum algorithm are facilitated on the basis, the data processing capacity of a quantum information algorithm model established on the basis of a quantum quick-calculation strategy and a quantum high-speed fuzzy hash method is improved remarkably, the calculation complex of the model is reduced remarkably, response delay is effectively reduced, the calculation complex of the algorithm is reduced, the response delay is reduced, and feature extraction and feature classification of the CTP financial data are achieved on the basis of the quantum information algorithm model.

The invention relates to a data transmission method, a data transmission device and a data transmission system. The data transmission method includes: establishing a TCP (transmission control protocol) control link between a network node and terminal equipment; receiving an HTTP (hyper text transfer protocol) request message sent by the terminal equipment via the control link; determining the quantity of data links required to be established according to a message header of a returned HTTP response message, and distributing data link port numbers and/or target IP (internet protocol) addresses for the data links required to be established; constructing a redirect message which is used for redirecting the HTTP request message to the data links; sending the redirect message to the terminal equipment via the control link; and receiving the HTTP request message resent by the terminal equipment via a first data link and sending the HTTP response message to the terminal equipment by the aid of the network node. By the aid of the data transmission method and the data transmission device, concurrent processing efficiency of data transmission and bandwidth availability ratio can be improved, response time delay of the response message is reduced, and network congestion is released.

The invention provides a technology that enables to control response delay of the supercharger in a variable compression ratio internal combustion engine under acceleration. When the variable compression ratio internal combustion engine is under acceleration, the compression ratio of the internal combustion engine is set to a compression ratio lower than a basic compression ratio that is so determined in accordance with the running condition of the internal combustion engine to realize a predetermined heat efficiency while suppressing knocking, thereby raising the exhaust gas pressure and shortening the response time of the supercharger.

In a vehicle control device, a brake feedback control unit calculates a brake feedback torque T_fb-BK using a PID control model when a current state is a brake feedback use state. The brake feedback control unit stops the calculation of the brake feedback torque T_fb-BK when the current state is a brake feedback limitation state, and stores as an output value the brake feedback torque T_fb-BK that is output at a brake limitation start timing. At a brake limitation release timing, the brake feedback control unit restarts the feedback control using the stored brake feedback torque T_fb-BK as an initial value. The brake mechanism can thereby generate a predetermined brake torque of not more than 0 [N·m] immediately after the brake limitation release timing.

Disclosed are a system and a method for distributing live broadcast content, wherein the method includes: a film library server converting a signaling sent by a live broadcast encoder for uploading/deleting/updating a file into live broadcast streaming data packets in a predetermined format; actively pushing or passively sending the converted live broadcast streaming data packets to an intermediate node server, or the intermediate node server receiving or actively pulling the live broadcast streaming data packets from the film library server until the live broadcast streaming data are actively pushed or passively sent to an edge node server; and the edge node server which receives the live broadcast streaming data packets recovering the live broadcast streaming data packets to the signaling previously sent by the live broadcast encoder, and generating/deleting/updating a corresponding media segment file and playlist description file according to an instruction of the signaling.

The invention discloses a public transport vehicle intersection priority pass control system and method based on edge computing, and the method comprises the steps: 1, receiving a bus priority pass demand, and judging whether the demand needs to carry out the priority pass control or not; 2, carrying out resource allocation on edge computing according to the requirements of priority pass control;3, obtaining a public transport vehicle intersection priority pass control scheme through edge computing; and 4, controlling the direction of the bus to be green and the conflict phase to be red according to the priority pass control scheme of the bus intersection. The edge computing technology is applied to intersection control, part or all of tasks originally executed in the cloud center can betransferred to the edge equipment to be executed, the pressure of the cloud center can be relieved, meanwhile, response time delay is reduced, data exposure is reduced, and external interference is reduced.

The invention provides a vehicle-mounted edge computing application caching method, device and system. The method comprises the steps of setting that n vehicle users exist in the coverage range of anedge server of a certain vehicle-mounted ad hoc network, and enabling each vehicle user i to request to unload a computing task of a vehicle-mounted application with the data size qi to the edge server; enabling the edge server to collect a calculation task of a vehicle-mounted application requested to be unloaded by a vehicle user within the coverage range of the edge server, and charge the corresponding vehicle user according to the data volume of the calculation task; after the calculation task is completed and a calculation result is obtained, feeding back the calculation result to a corresponding vehicle user, and selectively caching calculation task related data, including the calculation result, of the vehicle-mounted application unloaded by the vehicle user ; and if the computing tasks of the vehicle-mounted applications requested to be unloaded by other vehicle users subsequently are cached locally, enabling the edge server to directly feed back corresponding calculation results to the corresponding vehicle users. According to the invention, the response delay of the vehicle-mounted application service can be effectively reduced.

A method for controlling an engine having at least one cylinder is provided. The engine includes an intake manifold and an outlet control device for controlling flow from the intake manifold into the cylinder. The method includes generating an operator command. The method further includes controlling flow into the cylinder by changing the outlet control device to control flow into the cylinder based on the operator command.

In a mobile communication system for carrying out packet transmission/reception between a mobile terminal having a power saving mode to reduce a power consumption and a radio base station, a packet type of a packet is judged, and a sleep timer value indicating a time duration since a last packet is transmitted/received until the power saving mode starts is set according to a judgement result. Then, at least one sleep timer for measuring an elapsed time since the packet is transmitted/received is activated, and the mobile terminal is controlled to execute the power saving mode at a timing at which the elapsed time measured by the at least one sleep timer exceeded the sleep timer value.

The invention discloses a data transmission method and a relative system, wherein the method comprises that S202, via an embedded at a data sender, embedding an expansion row attach data with special format at the space outside the space occupied by voice data in a row attach data space, to transmit pure load data of expansion attach data to a data receiver, S204, a de-embed device at the data receiver de-embed the pure load data in the row attach data, to receive the pure load data in the expanded row attach data. The invention can fully utilize idle space to improve the utilization of system bandwidth.