Patents
Literature
Hiro is an intelligent assistant for R&D personnel, combined with Patent DNA, to facilitate innovative research.
Hiro

39results about How to "Reduce CPU usage" patented technology

Video producing method, device and system

The invention belongs to the technical field of video production, and particularly relates to a video producing method, device and system. The video producing method comprises the following steps: reading a configuration file, and acquiring and loading resources for producing a video; starting a rendering engine, and initializing a hardware encoding-decoding environment and an OpenGL off-screen rendering environment; and performing frame-by-frame off-screen rendering to generate raw video streams. Through adoption of an off-screen rendering technology, video production can be run in the form of a background task without being limited by the screen resolution. Through frame-by-frame rendering based on decoding time, audio-video synchronization and animation smoothness of an output video are ensured. Moreover, hardware decoding is performed on videos supporting hardware decoding, and hardware screen capture is performed on content which is already rendered in an off-screen way, so that hardware resources are utilized fully; the CPU (Central Processing Unit) occupation rate in a video producing process is lowered; and the processing efficiency is increased.
Owner:FUJIAN STAR NET EVIDEO INFORMATION SYST CO LTD

Display control method and system

The invention discloses a display control method and system. The display control method comprises the steps that a playing window for playing a first video is displayed on the top layer of a display interface of a terminal device, and the lower layer of the playing window is a video list page; a first operation instruction for displaying the video list page is received; the first operation instruction is responded to, the playing window is shrunk according to a preset scale and then is locked at the first position of the display interface, and the video list page is displayed at the second position of the current display interface. By the adoption of the display control method and system, the video can be clearly displayed, and the multiple-video information can be shown at the same time; in addition, the storage resources of the terminal device and the utilization rate of a CPU are reduced, and operation of equipment is smoother.
Owner:LETV INFORMATION TECH BEIJING

Attack detection method and device

The embodiments of the invention disclose an attack detection method and device; the method comprises the following steps: firstly, acquiring a target request message; extracting a to-be-detected field from the target request message; then, judging whether a firewall rule matched with the to-be-detected field exists in a blacklist rule base based on a pattern matching algorithm; if yes, detectingthe to-be-detected field based on a regular expression and determining whether the target request message is offensive or not; if not, directly judging that the target request message is not offensive. According to the scheme provided by the invention, for a normal legal request message, when attack detection is carried out, the CPU occupancy rate and the system performance consumption of a firewall system can be greatly reduced; moreover, most of the messages received by the firewall system are legal, so according to the embodiments of the invention, the working efficiency of the firewall system can be greatly improved.
Owner:BEIJING KINGSOFT CLOUD NETWORK TECH CO LTD +1

Robot navigation method, robot and storage device

The present invention discloses a robot navigation method and devices and belongs to the field of robot technologies. The method includes the following steps that: a global path from a starting pointto a target point is obtained through global cost map planning; a robot moves to the target point according to the global path, and performs path tracking during the moving process; whether a movementabnormality occurs during the path tracking is judged; if the movement abnormality occurs, a global cost map is used to re-plan a global path from the current location to the target point. Therefore,with the robot navigation method and devices of the invention adopted, the calculation quantity of the processor of a navigation device can be decreased, and navigation efficiency can be improved.
Owner:UBTECH ROBOTICS CORP LTD

Statistical method for realtime business information

A method for taking statistic to real-time service information, belongs to data communication technology field, this method is: the master control board of multi service exchanger send out inquiry command to each service board; after the service board receives the command, collect on-line users information about current service board, report to the master control board; after the master control board receives data package of first user, process the single user information; the service board reports the numbers of users to master control board, the master control board sets the service board as the state waiting for receiving the total numbers of known users; the master control board compares the total number of on-line users with the numbers of on-line users of this service board, if they are equal, then set the service board as state receiving all the information, when all the service board enter into the state receiving all the information , the master control board processes all the data, and output the result to the maintenance end. Using this method, CPU occupancy rate of master control board is low, doesn't effect the present service of service board.
Owner:HUAWEI TECH CO LTD

Cloud terminal decoder and load equalization algorithm thereof and decoding algorithm of GPU (Graphics Processing Unit)

The invention relates to the technical field of video data processing, particularly to a cloud terminal decoder of an MPEG2 (Moving Picture Experts Group) and a load equalization algorithm thereof and a decoding algorithm of a GPU (Graphics Processing Unit). The compression of the video codes is executed by a CPU (Central Processing Unit) and a GPU in the invention, namely, the motion estimation of the codes is implemented by the GPU. A feedback path between a rebuilt frame and the motion estimation is cut off. The rebuilt frame is replaced by an original frame for executing the motion estimation to eliminate the dependence between the CPU and GPU. The decoder disclosed by the invention effectively overcomes the problem of unsmooth video playing process caused by a high occupying rate of CPU and bottleneck resources when a plurality of cloud terminals play video at the same time, so that the decoder can be widely used in video data processing of cloud terminals.
Owner:G CLOUD TECH

Standard cell library function testing method with contrasting function

The invention relates to a standard cell library function testing method with a contrasting function. An input vector selection and extension module, a standard unit module, a contrasting unit module, an address selection module, a first output vector selection module, a second output vector selection module and a contrasting judgment module are arranged. The input vector selection and extension module is used for receiving or generating an initial vector and conducting selection and extension on the initial vector; the standard unit module stores all units in a standard cell library in a sorted mode; the contrasting unit module stores all standard cell ideal models in a sorted mode; the address selection module is used for generating a testing address and achieving the control function on other modules; the first output vector selection module and the second output vector selection module are used for receiving model output results of the standard units and an ideal cell respectively; the contrasting judgment module is used for contrasting the ideal output result with the actual output result. According to the standard cell library function testing method with the contrasting function, parallel testing is conducted on the multiple standard cells through vector reuse, redundant vectors are reduced, and the testing rate is increased; when functions of the standard cells are tested, contrasting of the testing results is directly conducted through an ideal model, whether each unit accords with the requirement or not is judged, and the workload of a tester is reduced.
Owner:BEIJING UNIV OF TECH

Cloud desktop system, image sequence compression code method and medium thereof

The invention discloses a cloud desktop system, an image sequence compression coding method and a medium thereof, which are used for at least reducing the CPU utilization rate of a server host. The system includes a server connected through a bus and at least one external device. The method is used for the server, and comprises the following steps: calling an external device to execute a preset pipeline parallel processing encoding process in a preset image encoding module; acquiring result data obtained after executing an encoding process from the external device; and calling the image encoding module to enable the image encoding module to complete the compression encoding of the cloud desktop image sequence according to the result data.
Owner:ZTE CORP

Processing method for exchanging system and continuous message

The system comprises separated managing module, exchanging module and a collection module. the collection module captures all message connection sent to the above exchanging module and saves it in buffer; according to whether or not reaching preset amount of messages or preset time delay it is decided if the message connection in buffer is sent to the exchanging module by using packet mode; the exchanging module receives and analyzes the packaged message connection, and then makes relevant treatment.
Owner:ZTE CORP

Processing method, device and system for generating chart component system

The embodiment of the invention discloses a processing method, device and system for generating a chart component system. The method comprises the steps of obtaining service structure information of a target chart; according to the service structure information, selecting a target component matched with the target chart from a preset chart component set; configuring the target component for generating a chart component system of the target chart. By utilizing the method or device, elements needed by constructing the chart can be modularized; due to the fact that components have the transportability, the components are used for constructing the chart component system, the extensibility mechanism of the system can be improved, and the coupling degree of the system can be lowered. Due to the component reusability, the memory size of the chart component system can be reduced; accordingly, the chart component system constructed by the components is applicable to a mobile terminal, and it is ensured that the CPU occupancy rate of the chart component system on the mobile terminal is low.
Owner:ADVANCED NEW TECH CO LTD

Flow control optimization method based on POTN

The invention discloses a flow control optimization method based on POTN. The method comprises the steps that a CCU application layer periodically sends a polling frame to each single disc, acquires in-place physical port information, and adds the physical port into or removes the physical port out of a flow control queue according to the in-place physical port information, to update the flow control queue in real time; and the CCU application layer sends data packets to every port in the flow control queue in a polling manner, and thus controls the flow of each port, wherein the packet sending time of each port and the maximal threshold of the total length of the data packets sent within the packet sending time are preset according to the packet sending rate of each port. According to the method, by dynamically maintaining the flow control queue, all ports in the flow control queue are actual in-place ports is ensured, unused ports on the device is prevented from adding into the flow control queue, the number of ports needing to be polled in the flow control queue is decreased, and thus the CPU occupancy rate of the flow control thread is significantly reduced.
Owner:FENGHUO COMM SCI & TECH CO LTD

Data transmitting and receiving methods and devices

The application discloses data transmitting and receiving methods and devices. One specific embodiment of the data transmitting method comprises the following steps: transmitting data to be transmitted; randomly generating an encryption length; encrypting the data from a predetermined position till an encryption length reaches the abovementioned encryption length in order to generated encrypted data; generating message header information including the encryption length; and constructing the message header information and the encrypted data into a complete message for transmission. Through the embodiment, the problem of high CPU (Central Processing Unit) occupation rate of a data transmission end is solved. One embodiment of the data receiving method comprises the following steps: acquiring received data; reading the message header information in the data, and extracting a decryption length; and decrypting the data from a predetermined position till a decryption length reaches the abovementioned decryption length in order to generate decrypted data. Through the embodiment, the problem of high CPU occupation rate of a data receiving end is solved.
Owner:BAIDU ONLINE NETWORK TECH (BEIJIBG) CO LTD

Running state information synchronizing method and device

The invention discloses a running state information synchronizing method and device. The method comprises the steps that a main virtual machine acquires current running state information of the main virtual machine when the main virtual machine confirms that the current running state meets the requirement of a running state information synchronizing strategy, and then the main virtual machine sends the running state information to an auxiliary virtual machine. According to the running state information synchronizing method and device, the CPU occupancy rate of a physical server can be reduced, processing performance of the physical server can be improved, and network consumption can be reduced.
Owner:NEW H3C TECH CO LTD

High-performance distributed container mirror image distribution system and method

The invention discloses a high-performance distributed container mirror image distribution system and method. The system comprises a main warehouse module, a scheduling module and a node processing module. The method comprises the steps that a main warehouse module forwards a mirror image pulling request; the scheduling module analyzes to obtain a to-be-pulled file list and sends data to the nodeprocessing module; the node processing module acquires the surprise hierarchical file and transmits the file through the IB network. According to the system and the method disclosed in the invention,the service pressure of the container mirror image server is reduced. The high-performance distributed container mirror image distribution system and method can be widely applied to the field of container cloud computing.
Owner:SUN YAT SEN UNIV

A method and apparatus for verifying an IEC 104 message

ActiveCN109167762AAvoiding Illegal Packet AttacksReduce CPU usageTransmissionComputer hardwarePrimary station
A method and apparatus for verifying an IEC 104 message are disclosed. The method for verifying IEC 104 message is characterized in that that device is applied to network equipment connecting IEC 104master station and IEC 104 slave station. The method comprises the follow steps: step A, judging whether the received message is an IEC 104 message when the message is received; If yes, the process goes to step B for processing; step B, verifying whether the message meets the preset IEC 104 message format requirements; If yes, the process is carried out in step C; step C, analyze that message, obtaining a plurality of preset characteristic information, and transferring the message and the obtained characteristic information to the step D for processing; step D, reading the pre-recorded sessionstate, and verifying whether the obtained characteristic information conforms to the preset session logic rule according to the read session state; If so, the message is judged to be legal.
Owner:HANGZHOU DPTECH TECH

AIO request processing method, system and device

The present application discloses an AIO request processing method, which comprises the following steps: receiving an AIO request input by a user; judging whether the AIO request is compatible with the underlying driver; if so, a SCSI request is constructed according to the AIO request, and the SCSI request is added into a request queue; constructing a SCSI command according to the SCSI request inthe request queue; the SCSI command being sent to the underlying driver so that the underlying driver executes the SCSI command to complete the processing of the AIO request. The technical proposal provided by the application omits the process of merging and sorting AIO and submitting the AIO to a special processing thread in the prior art, avoids the overhead of thread switching and scheduling algorithm, reduces the CPU occupancy, and reduces the processing delay of AIO requests. The present application also provides a system, a device and a computer-readable storage medium for AIO request processing, which has the above-mentioned beneficial effects.
Owner:ZHENGZHOU YUNHAI INFORMATION TECH CO LTD

RAID 6 hardware acceleration circuit structure and RAID 6 hardware acceleration implementation method

The invention discloses an RAID 6 hardware acceleration circuit structure, which comprises: a parameter analysis module, which is configured to be used for acquiring a parameter coefficient of an RAID 6 algorithm and calculating an operation coefficient based on the parameter coefficient; a data caching module, which is configured to be used for storing disk data; and a finite field operation module, which is configured to be used for sequentially obtaining the operation coefficient and the disk data corresponding to the current disk, and performing finite field multiplication on the basis of the operation coefficient and the disk data corresponding to the current disk to obtain an intermediate value of the current disk, wherein the finite field operation module is further configured to be used for caching the calculated intermediate value of the first disk to the intermediate value storage module, performing XOR operation on the calculated intermediate values of other disks and values in the intermediate value storage module, caching an operation result to the intermediate value storage module, and if the current disk is the last disk, outputting the value in the intermediate value storage module as a final result. The invention further discloses an RAID 6 hardware acceleration implementation method.
Owner:SUZHOU LANGCHAO INTELLIGENT TECH CO LTD

High-speed Ethernet frame reconstruction system based on FPGA

The invention discloses a high-speed Ethernet frame reconstruction system based on an FPGA (Field Programmable Gate Array), which is realized through the FPGA and comprises an Ethernet receiving and analyzing module, a packet analyzing module, a frame reconstruction module, a DDR and a data packet uploading module. the Ethernet receiving and analyzing module is used for receiving the electric signal subjected to photoelectric conversion from the optical port, analyzing and outputting an Ethernet frame data stream and a control signal through a physical layer and a data link layer, and sendingthe Ethernet frame data stream and the control signal to the packet analyzing module; the packet analysis module is used for sending the Ethernet frame data stream to the frame reconstruction module,and is also used for analyzing packet information from the Ethernet frame data stream according to the control signal and sending the packet information to the frame reconstruction module; the frame reconstruction module is used for logically synthesizing the parsed packet information and Ethernet frame data flow, reconstructing the parsed packet information and Ethernet frame data flow into a newdata frame and inputting the new data frame into DDR for storage; the DDR is used for caching data frames; the data packet uploading module is used for acquiring the data frame from the DDR and sending the data frame to an upper computer.
Owner:INST OF ACOUSTICS CHINESE ACAD OF SCI +1

Business processing method and system for shared memory

The embodiment of the invention relates to the technical field of communication, and discloses a service processing method and system for a shared memory, and the method comprises that: a domain management server extracts target service data from a service database, and stores the target service data into the shared memory; the domain management server generates a data file according to the target business data in the shared memory; and the domain management server distributes the data file to an application server according to a BT protocol, so that the application server performs service processing according to the data file. By means of the mode, the CPU occupancy rate of the database server is reduced, and the data loading efficiency is improved.
Owner:CHINA MOBILE GROUP ZHEJIANG +1

Refined cost management system based on business and financial integration for hospital in DRG payment mode

The invention discloses a refinement cost management system based on business and financial integration for a hospital in a DRG payment mode, and the system comprises: an application layer which is used for providing a user-oriented application-level service, and comprises a department cost analysis subsystem, a project cost analysis subsystem, and a disease cost analysis subsystem; a service layer whcih is used for providing basic support service for the application-level service, wherein the service layer comprises a sharing exchange system, a data center, a data intelligent analysis system and an application building system; and a resource layer which is used for providing hardware resources and data resources for the application layer and the service layer, wherein the hardware resources at least comprise a server host, storage equipment and network equipment, and the data resources comprise hospital business data and hospital financial data. According to the invention, a hospital business system and a financial system can be effectively connected, interconnection and intercommunication of business data and financial data are realized, and multi-dimensional data support is provided for comprehensive economic operation analysis of hospitals.
Owner:倪鑫

Data migration method and device and PCIE system

The invention provides a data migration method and device and a PCIE system. The data migration method applied to slave equipment comprises the following steps: configuring a memory space of the slave equipment; establishing mapping between the memory space of the slave equipment and a memory space of host equipment, and forming a first shared memory between the slave equipment and the host equipment, with the first shared memory being at least used for transmitting address information; and mapping the address information of the first shared memory in an outward mapping manner so as to execute data migration from the host equipment to the slave equipment and from the slave equipment to the host equipment. Data migration between the host equipment and the slave equipment is optimized, the CPU occupancy rate of the host equipment is reduced, the expansibility of the whole system is improved, and meanwhile the transmission performance between the host equipment and the slave equipment is improved.
Owner:SUZHOU KEDA TECH

Method and device for realizing docker network penetration through two-layer protocol stack in cloud computing environment

ActiveCN109309663BReduce forwarding speedReduce in quantityTransmissionData packNetworking protocol
The invention discloses a method and apparatus for enabling a docker network to penetrate through a two-layer protocol stack in a cloud computing environment. The method comprises the following steps:storing, by a docker network protocol stack, tcp / ip / eth related packet data and corresponding packet header information in a protocol stack cache for transparent transmission pre-created by a physical node, constructing a new data packet according to an address and a size corresponding to the cache, placing the new data packet in a related protocol stack packet processing function to transmit thesame to a cloud host protocol stack, so that the cloud host protocol stack judges whether the received data packet is placed in the protocol stack cache for transparent transmission, and if so, directly sending the received data packet to a physical node protocol stack; and upon receiving the data packet, extracting, by the physical node protocol stack, the address, the size and the correspondingpacket header information in the data packet, extracting the data stored in the protocol stack cache for transparent transmission according to the extracted address and the size, transmitting the data to a function corresponding to tcp, ip and eth layers to perform packet encapsulation processing, and sending the data to a physical network card for transmission. The number of data packets of thenetwork protocol stacks in the docker and a cloud host can be reduced, and the cpu usage rate and the data packet forwarding speed are reduced.
Owner:XIAMEN JIWEI TECH CO LTD

Method and system for redirecting audios and videos of HTML5 Video label of browser

The invention discloses a browser HTML5 Video label audio and video redirection method and system, and the method comprises the steps: intercepting the audio and video data inputted for the audio and video decoding class of a target video in a browser in a cloud virtual machine in a process that the browser plays the target video needing to be redirected; and the intercepted audio and video data are sent to the cloud terminal so as to be decoded, played and output through a player in the cloud terminal. According to the method, the CPU occupancy rate of the video played by the browser of the cloud virtual machine can be effectively reduced, the CPU consumption of compressed coding of the cloud server is also reduced, and meanwhile, the problem that the definition and fluency of the cloud desktop are reduced when the video is played by the browser is also solved.
Owner:湖南麒麟信安科技股份有限公司

The method of dynamic loading of c3 game map

The invention provides a C3 game map dynamic loading method. The method comprises the steps of 1, establishing a cell object, 2, cutting the landform mask data of game maps and scene and article data attached to basic unit cell data of each map in a parallel mode along four sides of the basic unit cell data, 3, further dividing the sub-map regions obtained after the cutting to obtain region units and storing all the landform mask data and the scene and article data in the region units in the data structure of a quadtree, 4, loading the landform mask data and the scene and article data of the cut diamond-shaped sub-map regions in a region unit loading mode when game map data need to be loaded for a game, and 5, carrying out permutation distribution after all the node region objects of the quadtree are loaded and accomplishing the loading of the game maps after all the sub-map regions of the maps are loaded. According to the C3 game map dynamic loading method, the usage rate of internal storage is reduced, the map loading efficiency of a mobile phone game system is improved, and the usage rate of a CPU is reduced.
Owner:FUJIAN TQ DIGITAL

Power supply fault reporting method

The invention provides a power supply fault reporting method. A substrate management controller generally does not access a fault register of a power supply unit, and only when the power supply unit sends a fault alarm signal, a field programmable gate array chip informs the substrate management controller that the power supply unit has a fault, and, at the moment, the BMC accesses the fault power supply unit through the baseboard management controller, collects fault information and reports a log. According to the power supply fault reporting method, the busy condition of the I2C bus connecting the substrate management controller and the power supply unit is greatly reduced, the I2C bus competition between the substrate management controller and the Intel management engine is reduced, and meanwhile the CPU utilization rate of the substrate management controller is reduced.
Owner:云尖信息技术有限公司

Address resolution method, cloud platform and medium

The invention discloses an address resolution method, a cloud platform and a medium. The method comprises the steps that virtual machine information is issued to an address resolution server through a control plane in the cloud platform; the virtual machine information comprises a mapping relation between IP addresses and MAC addresses of all virtual machines in the cloud platform and current position information; obtaining an ARP request message sent by a virtual machine in a local node through an address resolution agent deployed in each computing node, packaging the ARP request message into an address resolution request message, and sending the address resolution request message to an address resolution server; an address resolution server analyzes an MAC address corresponding to an IP address requested to be analyzed and current position information based on virtual machine information to obtain an analyzed MAC address and target position information, packages an address resolution response message based on the analyzed MAC address and the target position information, and returns the address resolution response message to a corresponding address resolution agent. The performance of the control plane can be improved, and thus the user experience is improved.
Owner:SHANDONG LANGCHAO YUNTOU INFORMATION TECH CO LTD

Video on-demand playing methods and devices based on articulated naturality web

PendingCN109547816ABroaden the application field of video transmissionReduce CPU usageSelective content distributionOccupancy rateProgress bar
The embodiment of the application provides video on-demand playing methods based on an articulated naturality web and corresponding devices. The method and the device are applied into the articulatednaturality web. An articulated naturality web server, an on-demand playing server and an articulated naturality web terminal in sequential communication connection are included in the articulated naturality web. Multiple pieces of video data are pre-stored in the on-demand playing server, and have video ID. The on-demand playing server generates a video playing list for the video ID, enables the articulated naturality web terminal to acquire the video data through the video playing list, and broadens the video transmission application field of the articulated naturality web. The on-demand playing server actively extracts corresponding video data from target video data according to a progress adjustment instruction of the articulated naturality web terminal, and then sends the same to the articulated naturality web terminal in real time, and enables the video data received by the articulated naturality web terminal after dragging a progress bar to be a real-time video stream. The videodata do not need to be pre-cached locally, and thus a CPU occupancy rate of playing the video data by the articulated naturality web terminal is reduced.
Owner:VISIONVERA INFORMATION TECH CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products