58results about How to "Low scalability" patented technology

A method and apparatus includes a plurality of processor groups each having a plurality of processor switch chips each having a plurality of processors and a processor crossbar, each processor connected to the processor crossbar; a plurality of switch groups each having a plurality of switch crossbar chips each having a plurality of switch groups each having a plurality of switch crossbar chips each having a plurality of switch crossbars each connected to a processor crossbar in each processor group, wherein no two switch crossbars in a switch group are connected to the same processor crossbar; a plurality of memory groups having a plurality of memory switch chips each having a plurality of memory controllers and a memory crossbar, each memory controller connected to the memory crossbar, each memory crossbar in each memory group connected to all of the switch crossbar in a corresponding one of the switch groups, wherein no two memory groups are connected to the same switch group.

The invention discloses a wireless positioning method and system. The wireless positioning system comprises at least one synchronous base station, a plurality of positioning base stations and at leastone positioning label and further comprises network communication equipment and a computing server. the synchronous base stations can also be used as positioning base stations at the same time, and the synchronous base stations and the positioning base stations further comprise high-precision clocks. The method comprises the following steps: a synchronization base station sends a synchronizationsignal at an interval of a period of time; the positioning base station receives the synchronization signal and uploads the data to the computing server; the computing server performs time synchronization on every two positioning base stations; whe positioning tag broadcasts a positioning signal once every other period of time; the positioning base station receives the positioning signal and uploads the data to the computing server; and the computing server computes the coordinates of the positioning tag.

A method, apparatus, and computer readable article of manufacture for tracing, on the basis of a key, an index created using a tree structure to access data. The method includes: searching the tree structure to reach a target node that can be associated with the key without acquiring a lock on the root node and on an internal node; setting the target node as a current node when the target node is a leaf node; moving the search to a node at a lower layer when the target node is an internal node by tracing an edge in the tree structure to reach a leaf node and setting the leaf node as the current node; and associating the key with the leaf node specified as the current node by acquiring a lock on the leaf node specified as the current node, thereby accessing the data.

The present invention provides methods and apparatuses for detecting neurons, that provide high sensitivity, low cost, durability, portability, and scalability. Neutrons interacting with a 10B layer in the present invention result in expression of alpha particles from the 10B layer. The alpha particles can then be detected, for example with a silicon photodetector or an imaging array (e.g., arrays used in digital cameras).

The invention relates to systems and methods for load balancing cloud-based accelerated transfer servers. The system may transfer target data from client computer systems into and / or out of cloud storage. The system may allocate a set of transfer server instances and provide a client computer system with a publicly available address of each of the allocated transfer server instances. The client computer system may transmit the target data to an allocated transfer server instance via the publicly available address using an accelerated network protocol that uses a connectionless network protocol. The allocated transfer server instance may cause the target data to be stored in cloud storage, such as a cloud object storage provided by a cloud service provider. A given transfer server instance may be integrated with the cloud service provider and have a local, low latency, connection to storage devices of the cloud service provider.

The present invention discloses a multiple fault coupling wheel case analog experimental table in the field of rotating machinery vibration monitoring and diagnosing, a plurality of typical faults of the gear equipment can be comprehensively, objectively and conveniently simulated, such as eccentricity, pitting, stripping, uniform abrasion on the tooth surface, tooth root fatigue crack, tooth breaking and the like; through dialing the assemble arrangement result of the gear with the fork spreader converting mechanism, the coupling fault signal and the signal varying process of the gear from normalization to fault generation can be continuously observed in operation. Besides, by arranging fault on the bearing the bearing fault and a bearing coupling fault of the gear can be simulated. The researching proofing procedure for the fault diagnosing method by the scientific researcher is greatly facilitated, the time from theory to the practice is shortened and the time and energy spent for arranging the experiment by the scientific researcher are saved. Besides, the multiple fault coupling wheel case analog experimental table can also be applied in the teaching practice for improving the learning and recognizing level.

The invention discloses an automatic electric quantity calibrating system and an automatic electric quantity calibrating method and belongs to the technical field of electromagnetism measuring. An automatic electric quantity calibrating device mainly comprises a computer, an AD / DA (analog to digital / digital to analog) module, a timer / counter module, a signal distribution module, a signal conditioning module, a signal transfer module, a system traceability measuring interface module, an equipment measuring interface module and a system power source module, and the automatic electric quantity calibrating system mainly comprises a self-checking system, a demarcating subsystem, an equipment measuring subsystem, a traceability subsystem and a data, certificate or report processing subsystem. The automatic electric quantity calibrating device has the advantages of small size, light weight, simplicity in operation, functional powerfulness, high accuracy and the like, the calibrating system applies signal-oriented designing technologies such as signal distribution, conditioning, converting and measurement adaption, measuring signal excitation and testing are realized, and requirements of automatized calibration of automatic testing systems under laboratory and carry-out conditions can be met.

The invention discloses an absorption film for eliminating bilirubin in blood and a preparation method thereof, and belongs to the technical field of biomedical materials. The method is characterized in that: a modifying method that chitosan is deposited on a hollow fiber film support layer and an outer surface of a hemodialyzer, and the chitosan is coupled with aglucon which can be interacted with the bilirubin is adopted; the coupled aglucon comprises polymine, polylysine, aminoethanol, 2,3-epoxypropyltrimethylammonium chloride, 3-chloro-2-hydroxypropltrimethyl ammonium chloride, betaine, lysine, arginine, dye Cibacron Blue F3GA, K-GRS blue, caprylaldehyde, n-butylamine, aniline, beta-cyclodextrin, bovine serum albumin and human sreum albumin. The method has the advantages of simple operation, mild condition, no damage to the structure of the film per se, capability of being coupled with various aglucons according to requirement and strong expansibility. When the absorption film is used for eliminating bilirubin, the bilirubin can be removed by the conventional dialysis with a simple system and low operation pressure.

The invention discloses an optical mechanical and electrical online measurement and control teaching innovation experimental platform and an experimental method thereof. The platform comprises a type C scanning mechanism and laser sensor part, a three-dimensional adjustment mechanism and CCD camera part, a tested object detection thin-aluminum plate platform part and a lower computer control and display part. The platform can be used for simulating the industrial field of online production of a tested object and online detection of the thickness and surface defects of the tested object, the knowledge points such as optical detection, mechanical control, single-chip control circuit, lower computer control and display, digital signal processing and implementation of software human-computer interface are combined, and the platform is suitable for students to simulate static and dynamic detection of the tested object under laboratory conditions and has the advantages of low cost, simulation performance, extensibility, comprehensiveness and innovativeness.

The invention discloses a wireless sensor network based micropower meter reading system and device. Each micropower wireless transmitting-receiving module is connected with intelligent electricity meters via RS485 wires, collects, packs and encodes power utilization information in the intelligent electricity meters, and transmits the information to a meter reading concentrator via a short range communication network; and the meter reading concentrator packs and encodes different packed data, in different periods, received from at least two micropower wireless transmitting-receiving modules, and sends the packed and encoded data to a data center via a remote communication network within prescribed time. A wireless sensor network media access control protocol is fused with a network clock synchronization mechanism, and the work state of an RF communication module is changed to improve the energy efficiency of the whole system on the basis that coordinated work and wakeup period of nodesin the network are controlled accurately.

The invention discloses an interconnection network system. The system is characterized in that routers in the system is divided into n router groups which are in complete connection, and each router group comprises one or more routers which are in complete connection. A self-adaptation routing method based on the novel interconnection network system is characterized in that the routers in the system are labeled, deadlock-free partial self-adaption routing can be achieved by special label decreasing direction priority algorithms or label increasing direction priority algorithms without using virtual passages, and the communication step length between optional processor nodes is not more than five. The invention further provides a complete self-adaptation routing method, deadlock-free complete self-adaptation routing can be achieved by utilizing the deadlock-free characteristics of the partial self-adaptation routing method aided by a novel flow control mechanism.

The invention discloses a time-sensitive network communication flow scheduling method based on deep reinforcement learning, and the method comprises the following steps: achieving the overall construction through the construction of a foundation, system modeling, a system framework, time slot selection, state modeling, motion modeling, environment modeling, and a deep neural network; detecting the whole body through error recovery, experimental demonstration and experimental objects, so the normal use of the whole body is ensured. According to the time-sensitive network communication flow scheduling method based on deep reinforcement learning, a DRL-based modeling, training and application method suitable for TS flow scheduling is provided, the method can be applied to different network environments by using different data training, and some optimization methods are provided; network information is expressed by using a directed graph, so that the DRLS can dynamically schedule the TS flow and can be quickly recovered when the network topology changes, and the uncertainty of the DNN is relieved and the scheduling capability and reliability of the scheduling method are improved by using a control gate technology.

The invention relates to a data transmission sharing method of a network topography system based on P2P, which is implemented by classifying nodes more precisely by testing the features of the nodes, and balancing the sharing performance of each node in the group; using the division of the interest area to reduce the network bandwidth extra consumption consumed by interaction of cache mapping information; reducing the scale of the divided nodes by a primary agency and a secondary agency and reducing the frequent change of node grouping caused by the change of interest area to reduce shaking effect during sharing; using tile distribution calculation and node service capability calculation and fully using the idle network resource of the nodes, improving node hit ratio and node match ratio during tile request by fast and evenly distributing the tiles, reducing the rate of sharing failure, thus improving the sharing efficiency. The invention provides a spatial data transmission and sharing scheme featuring low cost, high performance, high utility and high expandability.

The system includes system to be evaluated, monitoring desk of the user, service node for receiving command from the user and to control operation of the testing node, testing node for automatically setting up load generator according to disposal information given out by the service node and to send load request to system to be tested. The present invention has provided a evaluation system and method which are simple, expandable and easy to use as comparing with original performance evaluation method.

A device, and a service registration and searching method of a mixed distribution overlay network are disclosed. The invention comprises an internet, a service requester, a service provider, a physical network layer, a fuzzy query supported friend group network layer, which is based on the flooding peer-to-peer network searching art and a positioning network, which applies the peer-to-peer network searching art based on a distributed hash table. The invention also comprises that the service provider in the physical network layer registers the service for a friend group in the friend group network layer, the friend group registers to a friend group leader in the positioning network layer, the friend group leader registers at a service parser in the positioning network layer, the service requester of the service parser provides information for the service provider, so as to set up a connection between the service requester and the service provider. The invention uses the established positioning network layer and the friend group network layer to organically combine a flooding discovering and random forwarding mechanism, which is based on complete randomization with a keyword registration and searching based method. The invention enhances distributed network quality and provides very favorable support for an upper service.

The invention discloses a continuous data storage method, which comprises the steps of connecting a user end with a continuous data protection server, adding a user group in the continuous data protection server, and obtaining a group member information list; monitoring a public interface of the user group to wait for connection of group members of the user group; establishing a copy monitoring thread; connecting the group members of the user group according to the group member information list, judging identification of the group members and adding the group members into a corresponding group member list; monitoring and obtaining the synchronous requests of the group members, and establishing a copy of the main thread for the group members. The continuous data storage method takes the advantages of single storage and a local area network to improve the efficiency of data access. The data is no longer stored in the server, and is stored on other member host machines of the user group in the form of a plurality of copies. The continuous data storage method not only avoids performance bottle neck of the server and reduces service cost, but also ensures safety of the data. Even if a hard disc of a local host machine is damaged, the data still can be recovered.

The invention discloses a distributed high-concurrency data collection and processing method and system in an Internet of Things environment. The method comprises the following steps of: ; the methodcomprises the following steps of: creating a process guide Server Bootstrap; the method comprises the following steps: initializing a thread group bossGroup and a thread group workerGroup of which thetypes are NioEventLoopGroup, and initializing a thread group BossGroup and a thread group workerGroup of which the types are NioEventLoopGroup; the method comprises the following steps of: starting adata transponder process guider of which the type is Bootstrap; the method comprises the following steps of: constructing a processing class which is inherited from a ChannelInbound Handler Adapter;and adding processors Handler for different protocols, constructing a main interceptor, and binding a process port by utilizing the data receiving guider constructed in the above steps to monitor datasent by hardware equipment and analyze and transmit the data to a message queue unit. According to the distributed high-concurrency data collection and processing method and system in the Internet ofThings environment, a Netty framework is adopted, and large-batch data collection is supported.

The invention relates to a solar wireless monitoring system, which comprises at least one camera, wherein the camera is respectively connected with a wireless emitting machine and a charging and discharging controller, the charging and discharging controller is respectively connected with the wireless emitting machine, a solar photovoltaic module and a storage battery pack, the solar wireless monitoring system also comprises a wireless receiving machine, the wireless receiving machine is connected with a center server, and the center server is respectively connected with a mobile monitor and a video work station. The solar wireless monitoring system has the beneficial effects that: 1, a large amount of wire distribution work can be avoided, the construction cost is saved, the integral cost is low, the ditch digging and the pipe burying are not needed, and the solar wireless monitoring system is particularly suitable for outdoor long-distance and decorated occasions; 2, the wire limitation can be eliminated, and the advantages of short installation period and maintenance convenience are realized; 3, the networking is flexible, the expandability is good, and terminal equipment can be flexibly moved in the use process; 4, the solar wireless monitoring system can be applied to the monitoring of mobile equipment; and 5, the solar wireless monitoring system works at low voltage, safety and environment protection are realized, and the traditional energy sources are not consumed.

The invention provides a battery pack housing with heat exchange function and a battery pack, and belongs to the technical field of batteries. The battery pack housing comprises a heat exchange substrate and multiple side plates which are fixed on the heat exchange substrate, wherein the heat exchange substrate and multiple side plates are basically enclosed in order to form a cavity for accommodating and fixing one or more battery modules; the heat exchange substrate is formed by assembling one or several heat exchange substrate units, and each heat exchange substrate unit has an integral structure and multiple first channels are arranged therein; each side plate has an integral structure and at least one second channel is arranged therein; side plates and one or several heat exchange substrate units of the heat exchange substrate are welded, so that the first channels and multiple second channels are used together for forming heat exchange circulating channels, wherein liquid inlets and liquid outlets of the heat exchange circulating channels are arranged on at least one side plate which is provided with the second channels. The battery pack housing has the advantages of good utilization rate of materials, low cost, good extendability, and good reliability.

The embodiment of the invention provides a password operation realization method and device based on encryption cards and a server. The method mainly comprises following steps of respectively calculating the password operation capabilities of all encryption cards and the password operation capability of a server CPU (Central Processing Unit); allocating the to-be-allocated password operation quantities to the encryption cards or the server CPU based on a set password operation scheduling strategy, wherein the allocation is carried out according to the password operation capabilities of all encryption cards, the password operation capability of the server CPU and the to-be-processed password operation quantities in the encryption cards and the server CPU; and operating the allocated password operations by the encryption cards or the server CPU. According to the embodiment of the invention, the performance bottleneck defect of independently using one encryption card is solved; the defects of wasting cost and low scalability resulted from using multiple encryption cards are avoided; efficient password operation services such as signature verification are provided; a technical method which is featured by low cost, high efficiency and convenient use is provided for a signature verification server; and the excellent cost performance is realized.

The invention discloses a device and a method for automatically identifying the power model of a driver. The device comprises at least one parallel to serial logic chip and an MCU (Microprogrammed Control Unit) chip; the parallel to serial logic chip comprises the parallel signal input end and the serial signal end; the serial signal end comprises the clock signal input end, the signal loading end and the data output end; the MCU chip comprises at least three I / O (Input / Output) interfaces which are corresponding to the serial signal end; the clock signal input end, the signal loading end and the data output end are in signal connection with the corresponding I / O interfaces which are arranged on the MCU chip respectively. The method comprises step 1, arranging the high electrical level and or the low electrical level which are or is matched with the power model of the driver at the parallel signal input end of the parallel to serial logic chip; step 2, performing automatic identification on an identification code of the power model of the driver through the MCU chip; step 3, automatically arranging related parameters of the driver according to the identification code of the power model through the MCU chip. According to the device and the method for automatically identifying the power model of the driver, the power model of the driver can be automatically identified without manual intervention, the consumed time is short, and the identification is accurate.

The embodiment of the invention discloses a digital asset management method capable of realizing anonymous transaction, and the method comprises the steps: packaging digital assets into digital asset information through a secret key system and an encryption algorithm, packaging the digital asset information into encryption information and verification information, and carrying out the chain for evidence storage. The owner of the digital asset or the buyer of the digital asset can obtain the digital asset information by generating a symmetric encryption (decryption) key. And the encrypted information is used as a public key of each digital asset to replace a traditional user name and is used as an account identifier of the centralized service platform for collecting user use data. The unique code of the digital asset packaged in the digital asset information serves as a private key of the digital asset, the ownership of the digital asset can be verified, and the ownership of the digital asset can be relieved through logout of the unique code of the digital asset. And the uplink information is not directly associated with the key pair of the user, so that the reality of the user and the privacy of the network identity are guaranteed.

The invention discloses a linear scanning camera machine device and a visual instruction innovative experiment method. The linear scanning camera machine device comprises a linear scanning camera acquisition device, wherein the linear scanning camera acquisition device is communicated with a computer device with computer graphical interface software and a programming interface, is supported and fixed by using a bracket, and can be moved and adjusted upwards and downwards; and a motor-driven caterpillar band and a control device are arranged in the camera acquisition range of the linear scanning camera acquisition device, and are connected with the linear scanning camera acquisition device. The visual instruction innovative experiment method of the machine device has the advantages of low cost, expandability, fundamentality and innovation, and the application of a linear camera detection system on a motion occasion can be displayed.

The invention provides a login method for an interactive network television system. The method comprises the following steps: initiating a login request to a main service navigation system by an IPTV terminal according to the URL address of the main service navigation system stored therein; requesting login authorization of the main service navigation system to a service management system; returning an authorization result by the service management system to the main service navigation system; carrying out load balance by the main service navigation system, redirecting the login URL address of the user to edge service navigation systems after load balance; and calling the interface of an IPTC terminal by the edge service navigation systems, and rewriting the URL address of the IPTV terminal as the new address of an edge server. The process of the invention is clear and reliable, each sub-system in the system has clear responsibility, and the process has good extensibility. The number of the edge service navigation systems depends on the number of users. The method can enhance the low-cost characteristics, the reliability and the extensibility of each part of the interactive network television system, and improve the core competence of the IPTV system.

The invention discloses an automatic driving environment condition determination method and device and a storage medium, and belongs to the technical field of vehicle engineering. The automatic driving environment condition determination method comprises the following steps: standardizing acquired road working condition data; performing three-dimensional coverage combination on the standardized road condition data to obtain a data string of a plurality of scene combinations; and simulating the data strings of the plurality of scene combinations to obtain environmental conditions correspondingto the data strings of the plurality of scene combinations. In the present application, by standardizing acquired road working condition data and performing three-dimensional coverage combination on the standardized road working condition data, the automatic driving environment condition determination method obtains the data strings of the multiple scene combinations through scene simulation, so that the low redundancy, repeatability and expansibility of the data strings of the scene combinations are guaranteed, and scene simulation can be performed on the data strings of the multiple scene combinations, reliable virtual representation of the environment working condition of automatic driving is completed, and the confidence degree of environment working condition determination is improved.

The invention is suitable for the field of digital encoding, and provides a digital parallel multiplying unit. The multiplying unit comprises an input end A, an input end B0...BP-2BP-1, an output end C and a calculating unit, wherein the input end A and the input end B0...BP-2BP-1 input into the calculating unit respectively, and output from the output end C of the calculating unit after calculation and processing; an output calculating formula of the calculating unit is shown as C=R(AB0+AB1*<d>+...+ABp-1*<d(p-1)>)modF(x); in the formula, R is a non-zero polynomial, x is a radical of an irreducible polynomial F(x), d is a segmentation length, p is a segment number, and mod F(x) refers to reduction, namely, modulus operation of an obtained result. The multiplying unit mentioned in the application has the advantages of low space complexity, small area, low power consumption, increase of the calculating efficiency and expansibility.

The invention provides a calculation method of minimum energy consumption in green cloud service providing, and aims to overcome the resource bottleneck problem of a mobile terminal. The method comprises the following steps: A, a process of constructing a keyword word library L; B, a word frequency classification process; C, resource deployment for a new user request sent by a mobile user due to mobile micro-learning; and D, minimum energy function construction for the new user request. According to the method, dynamic TF-IDF is adopted for text classification, resources with high correct rates are placed on local cloud, resources with low correct rates are placed on public cloud, a two-layer cloud architecture model is constructed, and deployment of the mobile micro-learning resources is completed; on the basis of network environment and device state characteristics of real-time changing, a grey wolf optimizer (GWO) is utilized to estimate the energy consumption per byte processed by a system in a current environment state; and finally, a green and high-efficient total energy consumption function is constructed through analyzing a relationship between the user request and two-layer cloud architecture service providing.