48results about How to "Improve access bandwidth" patented technology

The invention relates to a large-scale distributed storage system in the technical field of computer network application, comprising at least one management node which is used for storing a name space, metadata information, an object number and address information of an object, a plurality of storage nodes which are used for storing corresponding object data and a plurality of super clients; the super clients access the management node by the storage node, search for the object number and the address information of the object of a file to be accessed in the management node and access the corresponding object data from the corresponding storage nodes; the management node comprises a file system which stores the name space and the metadata information of the object and a database which stores the object number and the address information of the object. Configurable automatic load balancing and data migration are adopted in the large-scale distributed storage system, the consistency of the data is periodically checked, and the junk data is periodically cleaned, therefore, various problems which are caused by manual operation are eliminated, and the large-scale distributed storage system has high expandability.

Embodiments of the present disclosure disclose a multi-interface solid state disk, and a processing method and system of the multi-interface solid state disk. The multi-interface solid state disk according to the present disclosure includes: plurality of interface control units, a command scheduling unit, a flash control unit and a flash chip. Each interface control unit corresponds to a communication interface respectively. The interface control unit receives an operating command through the communication interface. The command scheduling unit obtains, according to a scheduling rule, operating commands from the plurality of interface control units, puts the operating commands in a command queue, takes an operating command from the command queue, and sends the operating command to the flash control unit. The flash control unit converts the operating command into a flash operating command to operate the flash chip.

The embodiment of the invention discloses a multi-interface solid state disk (SSD) as well as a processing method and a system thereof and relates to the field of communication. A plurality of host computers directly share and access an SSD, and the requirements of all the host computers on high performance, high bandwidth and large volume of the SSD are met. The multi-interface SSD comprises a plurality of interface control units, a command dispatching unit, a flash control unit and a flash chip, wherein each interface control unit corresponds to a communication interface; the interface control units receive operating commands through the communication interfaces; the command dispatching unit obtains the operating commands from the plurality of interface control units according to a dispatching rule, places the operating commands into a command queue, and takes out the operating commands to be transmitted to the flash control unit; and the flash control unit converts the operating commands into flash operating commands for operating the flash chip.

The invention discloses a hybrid cloud storage method based on multi-level cache. The method comprises the following steps: a user joins a user group for login according to the local area network where the user is, and sends a user name and a password for user authentication to a super node of the region where the user group is, the super node achieves authentication of the user according to the received user name and password, the user receives a catalog view returned from the super node and metadata information of all files and sends the catalog view to a group leader node of the user group, the group leader node judges whether the buffer memory space of the group contains files in the catalog view according to the received catalog view, and if yes, feeds the IP address of the node in which the files are saved in the buffer memory space of the group to the user, and the user determines the optimal access path of each file in the catalog view according to the received feedback information. The hybrid cloud storage method allows the user to access data from the user group cache and the super node layer cache which have lower network delay, thereby lowering the network delay of the user data access.

The invention discloses a multi-core multi-thread microprocessor-oriented virtual active page buffer method and device. The method comprises the following steps of: constructing a virtual active page buffer; managing construction, memory and deletion of a virtual active page in the virtual active page buffer according to an access memory request of an on-chip access memory component of a processor; and directly reading the virtual active page from the virtual active page buffer to improve the access memory performance when the access memory request is a read request and hit occurs in the virtual active page buffer. The device comprises a page predictor (1), an arbitrator (2), a virtual active page mark array (3), a virtual active page data array (4), a read-write queue (5), a command queue (6), a data return queue (7), a pending table (8) and a multi-path selector (9). According to the method and the device, the integral active page number of the memory can be increased, the cost for re-excitation after closing of the active page is reduced, and the access memory bandwidth is increased; and the method and the device have the advantages of high access memory performance and high extensibility.

The present invention discloses multiple network equipment trunking implementation method. Said method includes extending partial terminal mark in calibrated link concentration into global terminal mark among multiple network equipment, selecting number of network ports in multiple network equipment, in each network equipment starting up trunking facilities having global terminal mark, configuring same IP and MAC address, same system priority and different equipment mark number address for virtual network port in each equipment, configuring same key value for said selected network port, using said selected network port as virtual network port slave equipment etc step; to congregate multiple network equipment network port. The present invention also can be realized by ASIC, and used as function module of switchboard and multiple function network card. The present invention is a IEEE 802.3 AD standard based specified trunking, being a important extending of standard trunking.

The invention provides a distributed network address translation system. The distributed network address translation system comprises an access layer, a multiplex layer and a control layer; middle nodes of the access layer are APs (Access Point); middle nodes of the multiplex layer are MPs (Multiplex Point); middle nodes of the control layer are CPs (Control Point); every AP is an access server; the functions of every AP comprise communicating with an internal network host, selecting public IP (Internet Protocol) resources for the internal network host, communicating with the MP owning the selected public IP resources and performing address translation; every MP is a multiplex server; the respectively owned public IP resources are maintained through every MP and meanwhile every MP is communicated with an external network through the respectively owned public IP resources; every CP comprises an IPS (IP Selector) and a CC (Control Center); every IPS is used for maintaining the public IP address resources of the MPs in the system; every CC is used for performing configuration on every layer of nodes; the APs, the MPs and the CPs are connected with each other through a tunnel which is a transmission layer connecting tunnel. The distributed network address translation system is used for performing the distributed network address translation and accordingly the utilization rate of the IP resources in the NAT (Network Address Translation) system can be improved.

The invention relates to the technical field of optical fiber-coaxial cable hybrid network access. Aiming at the limitations of the existing EoC technology application and combining with EPON technology application, the invention provides a multi-point control protocol based Ethernet passive electrical network (Ethernet Passive Electronic Network-EPEN), which comprises head end equipment connected with the FTTx optical node of an Ethernet passive optical network, terminal equipment (ENU) connected with user units and a passive electric distribution network (EDN), wherein the EDN consists of coaxial cables and one or more passive power splitters and provides electric channel connection between the ELT and the ENU, and a physical layer adopts the orthogonal frequency division multiplexing technology (OFDM); and a data link layer adopts the multi-point control protocol (MPCP), so that the two-way Ethernet data services carry in a point-to-multi-point coaxial access network. The invention can effectively realize the two-way transformation of an HFC network to which the EPON and the EPEN access in a compositing way, and has improved technology advancement, user access bandwidth and supported maximum number of users compared with other EoC technical standards.

The invention provides a solid state disk and an access control method thereof, and a solid state disk controller. The solid state disk controller comprises a host interface unit, a buffer unit, a storage interface unit and a compression / decompression unit, wherein the host interface unit is connected to a host, receives data from the host and sends data to the host; the buffer unit is connected to the host interface unit, and provides buffer for data transmission between the host interface unit and the compression / decompression unit; the compression / decompression unit is connected to the buffer unit, compresses data transmitted by the host to a storage unit, and decompresses data transmitted by the storage unit to the host; and one end of the storage interface unit is connected to the compression / decompression unit, while the other end is connected to the storage unit of the solid state disk for read-write operation on the storage unit. The effective capacity of the solid state disk is increased under the condition that the physical capacity of the solid state disk is fixed.

The invention discloses a method for designing an open type fused EOC. The open type fused EOC comprises a high-frequency submodule, a low-frequency submodule, an ONU submodule and the like. An IP data signal is modulated into a high-frequency signal and a low-frequency signal through the high-frequency submodule and the low-frequency submodule respectively after entering an open type fused EOC device through the ONU submodule, and the high-frequency signal, the low-frequency signal and a CATV signal are transmitted to a user side through the same cable. The open type fused EOC integrates the technical advantages of a high-frequency module, a low-frequency module and other modules, and fuses high-frequency technical standards and low-frequency technical standards, as a result, equipment cost is reduced, the access bandwidth is increased, the number of access users is increased, design work at an earlier stage and project implementation difficulty at a middle stage are reduced, operation and maintenance management, expansion and upgrading at a later stage are facilitated, and the open type fused EOC is applicable to the situation that distance between a photoreceiver and a user is different and other complex HFC network environments.

The invention of large-capacity high-speed cache memory comprises the high-speed cache memory controller and the high-speed cache memory units composed of DDR SDRAM array, and the address of the DDR SDRAM memory space changes in turn according array sequence. The invention uses DDR SDRAM memory array as a high-speed interface of cache memory, and provides greater access bandwidth and larger cache memory capacity for the high-speed interface, and also provides a relatively small delay for read and write access of the high-speed interface.

The invention belongs to the technical field of digital video encoding and decoding, in particular relates to a mapping method of video data of a video processor and memory space of a memory which includes six steps. According to the mapping relation, the characters of concurrency, etc. of data access in different pages of the memory with a page structure are fully utilized. The luminance component and the chromaticity component of video images are divided, and mapped into the corresponding page address, row address and column address of the memory in the modes of luminance grouping and chromaticity grouping, so that the access bandwidth of the video processor to the video data is improved to a great extent. The memory, replacing and update processes inside the memory of the video images entering the memory are prescribed, thereby guaranteeing the normal operation of the video processor.

The invention provides a multichannel RAM (random-access memory) reading / writing circuit and method. A plurality of RAMs are added in the circuit, a bus bridge, an arbiter and multiplexer cooperates, simpler circuit control logic is used to allow access bandwidth of RAM is increased greatly while system clock frequency is unchanged, and average system power consumption is unchanged. The multichannel RAM reading / writing circuit is simple, low in cost and easy to implement in a system.

The present invention discloses a software defined optical access network-based method and system for time-based bandwidth customization. The system comprises an application layer, a control plane and a data plane. The application layer receives and analyzes a bandwidth customization request of a user, and sends the bandwidth customization request through a northbound interface. The control plane receives and analyzes the bandwidth customization request, reserves network bandwidth resources according to a fact that a network bandwidth margin meets the bandwidth customization request, updates a bandwidth calendar, writes a to-be-delivered flow table into a node flow table and updates the node flow table, determines a flow table delivery time and starts a timer, reads the updated flow table when the time arrives, and sends an open flow message through a southbound interface. The data plane receives and analyzes the open flow message, sends the open flow message to a data abstraction layer according to the effective time of the open flow message, and completes configuration from the flow table to a forwarding device. Through adoption of the software defined optical access network-based method and system for time-based bandwidth customization, the access bandwidth is open to the user and is customized according to needs, actual network demands are met so as to improve the use experience, a network operator can develop a value added service, and the network flow can be smoothed by referring to a reasonable pricing policy.

The invention discloses a passive optical network, which comprises an optical line terminal, an optical cable and an optical network unit, wherein the optical line terminal comprises n transmitting light assembly units and m receiving light assembly units, and n and m are natural numbers of more than 1; and one end of the optical cable is connected with the optical line terminal, the other end of the optical cable is connected with an arrayed waveguide grating, and the arrayed waveguide grating is connected with the optical network unit. The passive optical network can realize point-to-point transmission of downlink optical signals so as to improve downlink access bandwidth of a terminal user.

The invention provides an afterglow generation method of a spectrum analyzer and the spectrum analyzer. The method comprises the steps that orthogonal mixing processing is carried out on an input signal of the spectrum analyzer to acquire a complex baseband signal; FFT calculation is carried out on the complex baseband signal to generate FFT calculation result data; according to preset integers Gand P, G FFT calculation results are in a group; P FFT calculation data in each FFT calculation result are in one piece, and FFT calculation result data are divided into A groups and B pieces; displayinterpolation processing is carried out on each FFT calculation result datum; the display interpolation processing result of each group in the same piece are subjected to histogram statistical accumulation to generate the histogram statistics of each piece of FFT calculation result data; and according to a predefined color mapping value and the histogram statistics of each piece in afterglow time, an afterglow display image is generated. According to the invention, the FFT calculation data are grouped and fragmented; the feature of high bandwidth efficiency of DRAM sequential access is fullyutilized; and a large number of calculations of display interpolation are completed in a short time.

An embodiment of the invention provides a storage device and an executive system and method of access operations. The device comprises an interlaced busbar, N memory banks and K direct-attached bus banks, wherein the letter K is smaller than or equal to the letter N, the interlaced busbar comprises M main interfaces and S auxiliary interfaces, the letter M is smaller than or equal to the letter S, and each of the M main interfaces is used for receiving continuous address access operations. Each of the N memory banks comprises S memories connected with the S auxiliary interfaces respectively. The interlaced busbar is used for encoding the continuous address access operations to at least one of the N memory banks by means of interlacing. The first one of the K direct-attached bus banks is used for receiving discrete address access operations and connecting the discrete address access operations to at least one of the N memory banks. Through the storage device and the executive system and method of access operations, the efficiency of access operations is guaranteed.

The invention discloses a method for storing a plurality of data in SDRAM, which comprises the following steps: each part of data is divided into a plurality of data blocks with the same size, the size of each data block is equal to the length of BURST which is used for reading the SDRAM internal storage data, and the data blocks of adjacent presetting part date are sequentially staggered and stored into a storing unit of an SDRAM storing array. In the method, the storing positions of the storing data in the SDRAM are arrayed, so the storage data in a SOC system which has larger probability of the concurrent access of a plurality of data packets by the similar offset in each MASTER can be more effectively accessed to save reading time, consequently, the access bandwidth of the SDRAM is increased, therefore, the performance of the whole system is improved.

The invention provides a construction method for the two-channel hybrid tunnel of mobile terminals. The construction method comprises the following steps: distributing the request information of a tunnel connecting IP address to a TAM sending request through pre-established TCP connection of a 3G channel with a TAM, and receiving the tunnel connecting IP address distributed by the TAM according to the request information; when a Wi-Fi signal verified by a wireless router is detected, establishing the TCP connection of a Wi-Fi channel with the TAM, and obtaining the amount of the mobile terminals connected with the TAM to establish the Wi-Fi channel; calculating the distribution ratio of data packets in the 3G channel and the Wi-Fi channel according to the channel parameters of the two channels and the amount of the mobile terminals, and storing and sending the distribution ratio to the TAM. A tunnel for using two physical channels to transmit data is established between the mobile terminals and the TAM, data transmission in the 3G channel is guaranteed all the time, and data transmission can be carried out at the same time in the Wi-Fi channel, so that the access bandwidths of the mobile terminals are improved, and data transmission delay caused by the switching of the channels is avoided.

The invention discloses a broadband acceleration management device. The device comprises an intelligent acceleration system, an operation supporting system and a bandwidth remote access server device. The intelligent acceleration system is connected with the operation supporting system through multiple data interfaces; and the operation supporting system comprises a network operation unit, a resource management unit, a service initiating unit, an automatic activating unit and a charging unit. According to the device, the integral access bandwidth of the user broadband can be simply and effectively increased; and the experience of all network application of a user can be increased and improved.

The invention relates to full-duplex radio-over-fiber systems. A system comprises a first link and a second link. The first link is used for radio-over-fiber of a W band from a first end to a second end. The first link comprises a first transmitting device and a first receiving device. The first transmitting device sets a central wavelength of a light source, generates a W band millimeter wave through utilization of a beat frequency, modulates a first original signal to the W band millimeter wave, obtains a first modulation signal and sends the first modulation signal to the first receiving device. The first receiving device is used for receiving and demodulating the firs modulation signal, thereby obtaining the first original signal. The second link is used for the radio-over-fiber of theW band from the second end to the first end. The second link and the first link are consistent in network topology structure. According to the scheme, the W band millimeter wave is generated throughutilization of the beat frequency, network configuration operation is simple, cost is greatly reduced, and relatively high access bandwidths are realized.

The invention relates to a passive optical interconnection network structure which comprises a two-stage optical switch. The two-stage optical switch receives the read / write request optical signals sent by N processing elements respectively, and exchanges according to the wavelength information in the read / write request optical signals so as to realize the non-blocking parallel access of the 16 processing elements, wherein the N is 4*4, 8*8, 16*16, 32*32, or 64*64. The invention relates to an on-chip passive optical interconnection network structure, is suitable for the communication of a large-scale multi-core processing element structure, and aims to use the wavelength division multiplexing technology to reduce the number of micro-ring resonators, improve the access bandwidth, and reducethe insertion loss and the access delay through the two-stage optical network switch structure.

The embodiment of the invention discloses a port access authentication method, a terminal and a storage medium, and the method comprises the steps: building an aggregation group of ports according tothe link configuration information of at least two ports; wherein one aggregation group corresponds to one aggregation port; creating a corresponding management port table entry for the aggregation group of the ports; obtaining an access authentication state of the aggregation port; and configuring the access authentication state of the aggregation port to a member port corresponding to the aggregation port according to the management port entry.

The invention relates to the technical field of networks / the technical field of wireless local area networks, and particularly relates to data acceleration technology and data detection technology on a WIFI network. The invention discloses a network device comprising a CPU, wherein the CPU is a multi-core CPU; the multi-core CPU comprises at least one basic core and at least one data task core; the basic core is used for managing and controlling the network device, managing and controlling the data task core and creating, managing or deleting a working process running in the data task core; and the data task core is used for running the working process in the data task core to process data flow corresponding to the network device. By means of the multi-core, acceleration processing is carried out on AP data to realize functions that cannot be realized by general AP, and thus the user experience is improved.

Embodiments of the invention disclose a storage unit. The storage unit comprises at least one single-bit storage structure, wherein each single-bit storage structure comprises four random access memories RAM: a first RAM, a second RAM, a third RAM and a fourth RAM, two read-out ports and two write-in ports; the two read-out ports comprise a first read-out port and a second read-out port; the two write-in ports comprise a first write-in port and a second write-in port; the first read-out port is connected with each of the first RAM and the second RAM, and is used for reading data stored in the first RAM and the second RAM; the second read-out port is connected with each of the third RAM and the fourth RAM, and is used for reading data stored in the third RAM and the fourth RAM; the first write-in port is connected with each of the first RAM and the third RAM, and is used for writing data into the first RAM and the third RAM; and the second write-in port is connected with each of the second RAM and the fourth RAM, and is used for writing data into the second RAM and the fourth RAM. Meanwhile, the embodiments of the invention furthermore disclose a processing system.

The invention provides a method for generating afterglow of a spectrum analyzer and the spectrum analyzer. The method includes: performing quadrature mixing processing on an input signal of the spectrum analyzer to obtain a complex baseband signal; performing FFT calculation on the complex baseband signal to generate FFT calculation result data; Preset integers G and P, with G times of FFT calculation results as a group, and P pieces of FFT calculation data in each FFT calculation result as one piece, the FFT calculation result data is divided into A group and B slice; for each FFT calculation result Perform display interpolation processing on the data; perform histogram statistical accumulation on the display interpolation processing results of each group in the same slice, and generate the histogram statistics of the FFT calculation result data of each slice; according to the predefined color mapping value and the histogram of each slice within the afterglow time Graph statistics generate persistence display images. The invention performs grouping and fragmentation processing on FFT calculation data, fully utilizes the characteristic of high bandwidth efficiency during sequential access of DRAM, and completes a large number of calculations of display interpolation in a relatively short time.

The invention provides a distributed network address translation system. The distributed network address translation system comprises an access layer, a multiplex layer and a control layer; middle nodes of the access layer are APs (Access Point); middle nodes of the multiplex layer are MPs (Multiplex Point); middle nodes of the control layer are CPs (Control Point); every AP is an access server; the functions of every AP comprise communicating with an internal network host, selecting public IP (Internet Protocol) resources for the internal network host, communicating with the MP owning the selected public IP resources and performing address translation; every MP is a multiplex server; the respectively owned public IP resources are maintained through every MP and meanwhile every MP is communicated with an external network through the respectively owned public IP resources; every CP comprises an IPS (IP Selector) and a CC (Control Center); every IPS is used for maintaining the public IP address resources of the MPs in the system; every CC is used for performing configuration on every layer of nodes; the APs, the MPs and the CPs are connected with each other through a tunnel which is a transmission layer connecting tunnel. The distributed network address translation system is used for performing the distributed network address translation and accordingly the utilization rate of the IP resources in the NAT (Network Address Translation) system can be improved.

The invention relates to a passive optical interconnection network structure. The passive optical interconnection network structure includes: two-stage optical switching switches; wherein, the two-stage optical switching switches respectively receive read / write request optical signals sent by N processing elements; according to the read / write request optical signals in the Wavelength information is exchanged to realize non-blocking parallel access among 16 processing elements; N is 4*4, 8*8, 16*16, 32*32 or 64*64. The invention relates to an on-chip passive optical interconnection network structure, which is suitable for communication of a large-scale multi-core processing unit structure, and aims to reduce the number of microring resonators and improve Access bandwidth, lower insertion loss, and lower access latency.

The invention discloses a network high available method and device based on a docker and relates to the technical field of a device or method for performing a special program. The method comprises the steps of setting virtual network card devices; adding a docker container network device; and bridging container network cards to the virtual network card devices through network bridge network cards. The method is based on a none network mode of the docker. The bridge network cards are added in a network plug-in mode. In a mode of linking the virtual network cards bonded with ports through the bridge network cards, various containers are supported to enjoy independent networks and independent IPs and do not share ports. Links are redundant, traffic loads are balanced, and the traffic bandwidth is increased. Through application of the method, the purchase cost of establishing a container virtualized system platform is reduced, and moreover, the reusability of server resources on operation of the same kinds of applications is improved.