WO2016086700A1 - Rack and communication method - Google Patents

Rack and communication method Download PDF

Info

Publication number
WO2016086700A1
WO2016086700A1 PCT/CN2015/089191 CN2015089191W WO2016086700A1 WO 2016086700 A1 WO2016086700 A1 WO 2016086700A1 CN 2015089191 W CN2015089191 W CN 2015089191W WO 2016086700 A1 WO2016086700 A1 WO 2016086700A1
Authority
WO
WIPO (PCT)
Prior art keywords
rack
computing
unit
network management
service
Prior art date
Application number
PCT/CN2015/089191
Other languages
French (fr)
Chinese (zh)
Inventor
姜亚星
Original Assignee
中兴通讯股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Publication of WO2016086700A1 publication Critical patent/WO2016086700A1/en

Links

Images

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04BTRANSMISSION
    • H04B10/00Transmission systems employing electromagnetic waves other than radio-waves, e.g. infrared, visible or ultraviolet light, or employing corpuscular radiation, e.g. quantum communication
    • H04B10/25Arrangements specific to fibre transmission

Definitions

  • the present invention relates to the field of optical communication technologies, and in particular, to a rack and a communication method.
  • Optical communication is communication in which light waves are used as carriers. It has the characteristics of large communication capacity, long relay distance, good confidentiality and low price.
  • optical communication technology has become the technology of choice for long-distance and large-capacity data transmission due to its huge bandwidth and transparency of transmission data.
  • the optical communication rack systems in the prior art use separate optical communication racks and supporting network management servers. In the mode, the two are connected by a network cable. The user command needs to go through the network management server to the rack control board and then to each board.
  • the specific process is shown in Figure 1.
  • Each board on the subrack uses an independent CPU design, which performs data processing separately and does not support distributed computing.
  • the embodiment of the invention provides a rack and a communication method, which realizes modular management and concurrent delivery of user commands, simplifies the installation and configuration of the equipment room, and can better meet the needs of the business change.
  • the embodiment of the present invention adopts the following technical solutions:
  • a rack includes: a rack housing, further comprising: being disposed in the first rack housing and connected to an external device through a backplane interface on the first rack housing Network management unit, disk array, and data processing unit, wherein
  • the network management unit is configured to process and format the user instructions, and then send the information to the service board, the data processing unit, and the disk array, and send the service status fed back by the service board to the user;
  • the data processing unit is configured to process data in the service board, the network management unit, and the disk array, and report the processing result of the data processing unit in real time;
  • the disk array is configured to store data in the network management unit and status information of each service card and log information of the service board.
  • the data processing unit includes: a computing unit overall controller located on the backboard, and a plurality of computing units connected to the computing unit total controller, where
  • the computing unit general controller is configured to record a computing capability and a CPU occupancy rate of the plurality of computing units, and allocate a calculation amount according to a remaining computing capability of the computing unit, wherein the remaining computing capability is a computing unit The difference between the total computing power and its CPU usage;
  • the calculation unit is configured to perform calculation processing on the data in the service board, the network management unit, and the disk array, and report the processing result of the calculation unit in real time through the calculation unit total controller.
  • the rack further includes: a service board disposed in the first rack casing and connected to the network management unit through a backplane interface on the first rack casing, where the service board is light
  • the signal is converted to a board of electrical signals and is configured to receive and transmit optical traffic signals.
  • the rack further includes: a second rack housing connected to the first rack housing, and a plurality of computing units and a plurality of service boards disposed in the second rack housing.
  • a communication method including:
  • the processing result and log information are stored.
  • the step of performing data processing according to the conversion instruction, and acquiring the processing result and the log information includes:
  • the remaining computing power is a difference between a total computing power of the computing unit and a CPU usage of the central processing unit;
  • the computing unit with the remaining computing power in all the computing units is selected to perform data processing on the conversion instruction, and the processing result and the log information are obtained.
  • the rack in the solution of the present invention discards the mode in which the network management server and the rack control board are separated in the original rack communication system, and separates the computing unit from the control board, so that the network management server and the control board can be combined into
  • the integrated network management unit in the solution of the present invention realizes direct control of the service board, and greatly improves the transmission efficiency of the user command transmission to the service board.
  • the computing resources of the entire rack are separated, and the data processing unit can process the data in the service board, the network management unit, and the disk array, thereby greatly improving the utilization of the computing resources of the rack and saving the cost.
  • the disk array on the rack in the solution of the present invention can be connected to the network management unit and the service card bus. Therefore, the network management unit and the service card can directly update and read data in the database, and greatly improve data transmission.
  • the transmission efficiency ensures the correctness of data transmission.
  • Figure 1 shows a connection diagram of a rack system in the prior art
  • Figure 3 is a flow chart showing the instructions issued in the prior art
  • Figure 4 is a flow chart showing the calculation of data in the embodiment of the present invention.
  • Figure 5 is a flow chart showing the calculation of data in the prior art
  • Figure 7 is a flow chart showing log management in the prior art
  • Figure 8 is a structural view showing an integrated rack in the embodiment of the present invention.
  • Figure 9 is a structural view showing a master-slave frame in an embodiment of the present invention.
  • Figure 10 is a flow chart showing the communication method in the embodiment of the present invention.
  • Embodiments of the present invention disclose a rack that includes a rack housing, and further includes: a network management device disposed in the first rack housing and connected to an external device through a backplane interface on the first rack housing Units, disk arrays, and data processing units,
  • the network management control unit is configured to process and format the user instructions, and then send the information to the service board, the data processing unit, and the disk array, and send the service status fed back by the service board to the user;
  • the network management unit integrates the network management server and the rack control board in the prior art, and the network management control The unit and the frame of the rack are connected through the backplane interface bus to realize parallel communication.
  • the network management unit can directly operate the service board, and can process and format the user instructions and send it to the service board and data processing unit.
  • the disk array sends the service status fed back by the service board to the user.
  • the transfer efficiency of the user instruction to the service board is greatly improved, and the network management server that needs to be additionally placed is avoided, which saves the steps of separately installing the network management server, and simplifies the process.
  • the site is installed and configured. At the same time, the interaction between the network management unit and the boards on the rack is improved, and the reliability and timely reflection capability of the rack are increased.
  • the network management software of the rack communication system can also be run on the network management unit to control the communication and data processing of the overall rack system, such as receiving and feedback of user commands and running. Processing of data in the process.
  • the data processing unit is configured to process data in the service board, the network management unit, and the disk array, and report the processing result of the data processing unit in real time;
  • the data processing unit includes: a calculation unit total controller located on the backboard and a plurality of calculation units connected to the calculation unit total controller, wherein
  • a calculation unit total controller configured to record a calculation capability and a CPU occupancy rate of the plurality of calculation units, and allocate a calculation amount according to a remaining calculation capability of the calculation unit, wherein the remaining calculation capability is a total calculation capability of the calculation unit and The difference in CPU usage;
  • the calculation unit is configured to perform calculation processing on the data in the service board, the network management unit, and the disk array, and report the processing result of the calculation unit in real time through the total controller of the calculation unit.
  • the data processing unit can process data in the service board, the network management unit, and the disk array, and adopt The distributed computing method can separate all computing resources of the entire rack, greatly improving the utilization of rack computing resources.
  • the computing unit overall controller in the embodiment of the present invention can record the computing power and CPU usage of multiple computing units, and in the prior art, each individual board
  • the use of independent CPU and memory avoids the fact that some boards in the prior art are saturated while other boards are idle, and the number of data processing cards can be increased or decreased according to user requirements, which saves cost and is convenient. Subsequent control and maintenance.
  • the data processing unit in the embodiment of the present invention also supports hot plugging, and data is not lost during the plugging and unplugging process, which increases the maintainability of the rack and the service life of the rack.
  • the disk array is configured to store data in the network management unit and status information of each service board and log information of the service board.
  • FIG. 6 which is a flowchart of the log management in the embodiment of the present invention
  • the service board can directly Update and read the data in the disk array, compared with the prior art log management process in Figure 7, avoiding the update and read operations, first through the control board and then to the network management server to the database communication mode, greatly improved Data transfer efficiency and guaranteed number According to the correctness of the transmission.
  • the disk array can also support hot swapping, and data will not be lost during the plugging process.
  • the disk array can also be used as the database of the network management software running on the network management unit, and is configured to store related data of the network management software and data in the network management unit.
  • the service card is a card that converts an optical signal into an electrical signal, and is configured to receive and transmit an optical service signal.
  • the service board is disposed in the first rack enclosure and connected to the gateway control unit through a backplane interface on the first rack enclosure.
  • the structural diagram of the integrated rack 800 that is, the network management unit, the data processing unit, the disk array, and the service board are all installed on the same rack, which is more convenient for the rack. Installation configuration.
  • a second rack may be added according to actual conditions.
  • the first rack shell and the second rack shell are connected through a data interface, and the second rack shell is further provided with a plurality of computing units and service boards.
  • the plurality of computing units and service boards are connected to the gateway control unit located in the first rack housing through an interface on the second rack housing and an interface on the first rack housing.
  • main frame 901 and the slave chassis 903 have the structural diagrams as shown in the figure
  • the main frame 901 and the slave chassis 903 transmit data through the data interfaces respectively located on the master and slave racks.
  • the computing unit on the mainframe is configured to process the data calculation of the mainframe web control unit, and the computing unit from the rack is set to calculate the data on the service board.
  • This master-slave rack configuration is more suitable for sites with large traffic handling.
  • the rack of the software-defined network SDN network architecture can be used as a slave rack, which can facilitate the control plane of the SDN network architecture to process service data.
  • a communication method is disclosed in the embodiment of the present invention. As shown in FIG. 10, the method 1000 includes the following steps:
  • Step S1001 Acquire a user instruction, and perform processing and format conversion according to a user instruction
  • Step S1003 Perform data processing according to the conversion instruction, and obtain processing results and log information
  • step S1003 specifically includes:
  • the calculation unit with the remaining computing power remaining in all the computing units is selected to perform data processing on the conversion instruction, and the processing result and the log information are obtained.
  • Step S1005 Store the processing result and the log information.
  • the network management unit receives the instruction sent by the user, queries the processing result of the data processing unit stored in the disk array, and the status information of each service card, and then sends the user instruction to the corresponding
  • the business board is processed and the results are reported back to the user.
  • the recorded logs are sent to the disk array for storage in real time.
  • the data processing unit calculates the data that needs to be calculated on the network management unit and the service board. When performing data calculation, the total calculation capability and CPU usage of all current computing units are first recorded by the calculation unit total controller.
  • the difference between the total computing power of the computing unit and the CPU usage is recorded as the remaining computing power of the computing unit, and then the computing unit with the strong computing power among all remaining computing units is selected, and the data calculation is prioritized.
  • the distributed metering method is mainly used to realize the balanced allocation of computing resources, avoiding the problem that computing resources cannot be shared, and greatly improving the utilization of computing resources.
  • modules or steps of the present invention described above can be implemented by a general-purpose computing device that can be centralized on a single computing device or distributed across a network of multiple computing devices. Alternatively, they may be implemented by program code executable by the computing device such that they may be stored in the storage device by the computing device and, in some cases, may be different from the order herein.
  • the steps shown or described are performed, or they are separately fabricated into individual integrated circuit modules, or a plurality of modules or steps thereof are fabricated as a single integrated circuit module.
  • the invention is not limited to any specific combination of hardware and software.
  • the rack and the communication method provided by the embodiments of the present invention have the following beneficial effects: modular management and concurrent delivery of user commands, simplifying the installation and configuration of the equipment room, and better meeting the change of the service. Demand.

Abstract

Embodiments of the present invention relate to the technical field of optical communications and provide a rack and a communication method. The rack comprises a rack housing, and further comprises a network management control unit, a disk array, and a data processing unit that are disposed in a first rack housing and that are connected to an external device through a backplane interface on the first rack housing. The network management control unit is configured to process and perform format conversion on a user instruction, and then send the user instruction to a service board card, the data processing unit, and the disk array, and send, to a user, a service state fed back by the service board card. The data processing unit is configured to process data in the service board card, the network management control unit, and the disk array, and report a processing result of the data processing unit in real time. The disk array is configured to store data in the network management control unit, state information of each service board card, and log information of the service board card. The embodiments of the present invention improve reliability and maintainability of the rack, and also improve a reaction capability and a service life of the rack.

Description

一种机架及通讯方法Rack and communication method 技术领域Technical field
本发明涉及光通讯技术领域,尤其涉及一种机架及通讯方法。The present invention relates to the field of optical communication technologies, and in particular, to a rack and a communication method.
背景技术Background technique
光通信就是以光波为载波的通信。其具有通信容量大、中继距离长、保密性好、价格低廉等特点。Optical communication is communication in which light waves are used as carriers. It has the characteristics of large communication capacity, long relay distance, good confidentiality and low price.
近年来,互联网技术飞速发展,互联网用户对海量数据、多样化的信息服务、复杂的数据聚合与交互的需求极大的促进了通讯网络的发展。其中,光通信技术以其所带来的巨大带宽和传输数据透明性等特点,成为长距离大容量数据传输的首选技术。In recent years, Internet technology has developed rapidly, and the demand for massive data, diversified information services, and complex data aggregation and interaction among Internet users has greatly promoted the development of communication networks. Among them, optical communication technology has become the technology of choice for long-distance and large-capacity data transmission due to its huge bandwidth and transparency of transmission data.
但当前光通讯机架却已陷了一种高成本、体积庞大但利用率不高的怪圈中:现有技术中的光通讯机架系统都采用分开的光通信机架和配套的网管服务器的方式,两者依靠网线连接,用户指令需要通过网管服务器再到机架控制单板再到各个板卡,具体流程如附图1所示。其中子架上每个板卡都采用独立CPU设计,单独进行数据处理,不支持分布式计算。However, the current optical communication rack has been trapped in a high-cost, bulky, but low-utilization cycle: the optical communication rack systems in the prior art use separate optical communication racks and supporting network management servers. In the mode, the two are connected by a network cable. The user command needs to go through the network management server to the rack control board and then to each board. The specific process is shown in Figure 1. Each board on the subrack uses an independent CPU design, which performs data processing separately and does not support distributed computing.
这种方式存在着通讯速度慢,CPU资源无法共享,需要另外放置网管服务器等缺点;而且目前使用的光通信板卡,每个板卡都使用独立的CPU和内存,可能出现部分业务板卡计算资源已经饱和,但其他业务板卡却一直空闲的状况,大大浪费已有的硬件资源,也不便于后续对机架计算能力的升级和维护操作。In this way, there are shortcomings such as slow communication speed, inability to share CPU resources, and additional network management server. Moreover, the optical communication boards currently used use separate CPUs and memory for each board, and some business card calculations may occur. The resources are saturated, but the other service boards have been idle, which greatly wastes the existing hardware resources and is not convenient for subsequent upgrade and maintenance operations on the rack computing power.
发明内容Summary of the invention
本发明实施例提供了一种机架及通讯方法,实现了模块化管理和对用户命令并行下发,简化了机房安装配置,能够更好地满足业务发生变化时的需求。 The embodiment of the invention provides a rack and a communication method, which realizes modular management and concurrent delivery of user commands, simplifies the installation and configuration of the equipment room, and can better meet the needs of the business change.
为了解决上述技术问题,本发明实施例采用如下技术方案:In order to solve the above technical problem, the embodiment of the present invention adopts the following technical solutions:
依据本发明实施例的一个方面,提供了一种机架,包括:机架外壳,还包括:设置于第一机架外壳内,并通过第一机架外壳上的背板接口与外部设备连接的网管控制单元、磁盘阵列以及数据处理单元,其中,According to an aspect of an embodiment of the present invention, a rack includes: a rack housing, further comprising: being disposed in the first rack housing and connected to an external device through a backplane interface on the first rack housing Network management unit, disk array, and data processing unit, wherein
所述网管控制单元,设置为将用户指令进行处理及格式转换后发送给业务板卡、数据处理单元、磁盘阵列,并将业务板卡反馈的业务状态发送给用户;The network management unit is configured to process and format the user instructions, and then send the information to the service board, the data processing unit, and the disk array, and send the service status fed back by the service board to the user;
所述数据处理单元,设置为对业务板卡、网管控制单元、磁盘阵列中的数据进行处理,并实时上报所述数据处理单元的处理结果;The data processing unit is configured to process data in the service board, the network management unit, and the disk array, and report the processing result of the data processing unit in real time;
所述磁盘阵列,设置为存储所述网管控制单元中的数据及各业务板卡的状态信息以及业务板卡的日志信息。The disk array is configured to store data in the network management unit and status information of each service card and log information of the service board.
可选地,所述数据处理单元包括:位于背板上的计算单元总控制器以及与所述计算单元总控器连接的多个计算单元,其中,Optionally, the data processing unit includes: a computing unit overall controller located on the backboard, and a plurality of computing units connected to the computing unit total controller, where
所述计算单元总控制器,设置为记录所述多个计算单元的计算能力及CPU占用率,并根据所述计算单元的剩余计算能力分配计算量,其中,所述剩余计算能力为计算单元的总计算能力与其CPU占用率的差值;The computing unit general controller is configured to record a computing capability and a CPU occupancy rate of the plurality of computing units, and allocate a calculation amount according to a remaining computing capability of the computing unit, wherein the remaining computing capability is a computing unit The difference between the total computing power and its CPU usage;
所述计算单元,设置为对业务板卡、网管控制单元、磁盘阵列中的数据进行计算处理,并通过计算单元总控制器实时上报所述计算单元的处理结果。The calculation unit is configured to perform calculation processing on the data in the service board, the network management unit, and the disk array, and report the processing result of the calculation unit in real time through the calculation unit total controller.
可选地,所述机架还包括:设置于第一机架外壳内,并通过第一机架外壳上的背板接口与网管控制单元连接的业务板卡,所述业务板卡为将光信号转化为电信号的板卡,并设置为接收和发送光业务信号。Optionally, the rack further includes: a service board disposed in the first rack casing and connected to the network management unit through a backplane interface on the first rack casing, where the service board is light The signal is converted to a board of electrical signals and is configured to receive and transmit optical traffic signals.
可选地了,所述机架还包括:与第一机架外壳连接的第二机架外壳,以及设置于所述第二机架外壳内的多个计算单元和多个业务板卡。Optionally, the rack further includes: a second rack housing connected to the first rack housing, and a plurality of computing units and a plurality of service boards disposed in the second rack housing.
依据本发明实施例的另一个方面,还提供了一种通讯方法,包括:According to another aspect of the embodiments of the present invention, a communication method is further provided, including:
获取用户指令,根据所述用户指令进行处理及格式转换;Obtaining a user instruction, performing processing and format conversion according to the user instruction;
根据所述转换指令进行数据处理,获取处理结果及日志信息; Performing data processing according to the conversion instruction, and obtaining processing results and log information;
将所述处理结果及日志信息进行存储。The processing result and log information are stored.
可选地,根据所述转换指令进行数据处理,获取处理结果及日志信息的步骤包括:Optionally, the step of performing data processing according to the conversion instruction, and acquiring the processing result and the log information includes:
获取所有计算单元的剩余计算能力,其中,所述剩余计算能力为计算单元的总计算能力与其中央处理器CPU占用率的差值;Obtaining a remaining computing power of all computing units, wherein the remaining computing power is a difference between a total computing power of the computing unit and a CPU usage of the central processing unit;
选择所有计算单元中所述剩余计算能力较强的计算单元对所述转换指令进行数据处理,获取处理结果及日志信息。The computing unit with the remaining computing power in all the computing units is selected to perform data processing on the conversion instruction, and the processing result and the log information are obtained.
本发明的有益效果是:The beneficial effects of the invention are:
本发明方案中的机架,摒弃了原机架通信系统中网管服务器和机架控制板卡分离的模式,将计算单元从控制板卡上分离出来,因此能够将网管服务器与控制板卡合为一体构成本发明方案中网管控制单元,实现了对业务板卡的直接控制,大大提升了用户指令传递到业务板卡的传递效率。且将整个机架的计算资源独立出来,数据处理单元能够对业务板卡、网管控制单元、磁盘阵列中的数据进行处理,大大提高了机架的计算资源利用率,节约了成本。另外,本发明方案中机架上的磁盘阵列能够与网管控制单元以及业务卡板总线连接,因此网管控制单元与业务卡板可以直接更新和读取数据库中的数据,并大大提升了数据传输的传递效率,保证了数据传输的正确性。The rack in the solution of the present invention discards the mode in which the network management server and the rack control board are separated in the original rack communication system, and separates the computing unit from the control board, so that the network management server and the control board can be combined into The integrated network management unit in the solution of the present invention realizes direct control of the service board, and greatly improves the transmission efficiency of the user command transmission to the service board. Moreover, the computing resources of the entire rack are separated, and the data processing unit can process the data in the service board, the network management unit, and the disk array, thereby greatly improving the utilization of the computing resources of the rack and saving the cost. In addition, the disk array on the rack in the solution of the present invention can be connected to the network management unit and the service card bus. Therefore, the network management unit and the service card can directly update and read data in the database, and greatly improve data transmission. The transmission efficiency ensures the correctness of data transmission.
附图说明DRAWINGS
图1表示现有技术中机架系统的连接图;Figure 1 shows a connection diagram of a rack system in the prior art;
图2表示本发明实施例中指令下发的流程图;2 is a flowchart showing an instruction issued in an embodiment of the present invention;
图3表示现有技术中指令下发的流程图;Figure 3 is a flow chart showing the instructions issued in the prior art;
图4表示本发明实施例中数据计算的流程图;Figure 4 is a flow chart showing the calculation of data in the embodiment of the present invention;
图5表示现有技术中数据计算的流程图;Figure 5 is a flow chart showing the calculation of data in the prior art;
图6表示本发明实施例中日志管理的流程图;6 is a flowchart of log management in an embodiment of the present invention;
图7表示现有技术中日志管理的流程图; Figure 7 is a flow chart showing log management in the prior art;
图8表示本发明实施例中一体化机架的结构图;Figure 8 is a structural view showing an integrated rack in the embodiment of the present invention;
图9表示本发明实施例中主从机架的结构图;Figure 9 is a structural view showing a master-slave frame in an embodiment of the present invention;
图10表示本发明实施例中通讯方法的流程图。Figure 10 is a flow chart showing the communication method in the embodiment of the present invention.
具体实施方式detailed description
为使本发明的目的、技术方案和优点更加清楚,下面将结合附图及具体实施例对本发明进行详细描述。The present invention will be described in detail below with reference to the drawings and specific embodiments.
实施例一Embodiment 1
本发明实施例公开了一种机架,该机架包括机架外壳,还包括:设置于第一机架外壳内,并通过第一机架外壳上的背板接口与外部设备连接的网管控制单元、磁盘阵列以及数据处理单元,Embodiments of the present invention disclose a rack that includes a rack housing, and further includes: a network management device disposed in the first rack housing and connected to an external device through a backplane interface on the first rack housing Units, disk arrays, and data processing units,
网管控制单元,设置为将用户指令进行处理及格式转换后发送给业务板卡、数据处理单元、磁盘阵列,并将业务板卡反馈的业务状态发送给用户;The network management control unit is configured to process and format the user instructions, and then send the information to the service board, the data processing unit, and the disk array, and send the service status fed back by the service board to the user;
具体地,如图2所示,为本发明实施例中指令下发的流程图,在本发明实施例中,网管控制单元融合了现有技术中的网管服务器与机架控制板,且网管控制单元与机架外壳通过背板接口总线连接,实现了并行通讯,网管控制单元能够直接对业务板卡进行操作,且能够将用户指令进行处理及格式转换后发送给业务板卡、数据处理单元、磁盘阵列,并将业务板卡反馈的业务状态发送给用户。与图3中现有技术指令下发的流程相比,大大提升了用户指令传递到业务板卡的传递效率,同时避免了需要额外放置的网管服务器,节省了单独安装网管服务器的步骤,简化了站点安装配置,同时,还提高了网管控制单元与机架上各单板之间信息的交互,增加了机架的可靠性和及时反映能力。Specifically, as shown in FIG. 2, which is a flowchart of a command issued in the embodiment of the present invention, in the embodiment of the present invention, the network management unit integrates the network management server and the rack control board in the prior art, and the network management control The unit and the frame of the rack are connected through the backplane interface bus to realize parallel communication. The network management unit can directly operate the service board, and can process and format the user instructions and send it to the service board and data processing unit. The disk array sends the service status fed back by the service board to the user. Compared with the process issued by the prior art instruction in FIG. 3, the transfer efficiency of the user instruction to the service board is greatly improved, and the network management server that needs to be additionally placed is avoided, which saves the steps of separately installing the network management server, and simplifies the process. The site is installed and configured. At the same time, the interaction between the network management unit and the boards on the rack is improved, and the reliability and timely reflection capability of the rack are increased.
进一步地,在本发明实施例中,还能够在网管控制单元上运行该机架通讯系统的网络管理软件,控制整体机架系统的通讯及数据的处理,如对用户指令的接收和反馈以及运行过程中数据的处理。 Further, in the embodiment of the present invention, the network management software of the rack communication system can also be run on the network management unit to control the communication and data processing of the overall rack system, such as receiving and feedback of user commands and running. Processing of data in the process.
数据处理单元,设置为对业务板卡、网管控制单元、磁盘阵列中的数据进行处理,并实时上报数据处理单元的处理结果;The data processing unit is configured to process data in the service board, the network management unit, and the disk array, and report the processing result of the data processing unit in real time;
其中数据处理单元包括:位于背板上的计算单元总控制器以及与计算单元总控器连接的多个计算单元,其中,The data processing unit includes: a calculation unit total controller located on the backboard and a plurality of calculation units connected to the calculation unit total controller, wherein
计算单元总控制器,设置为记录多个计算单元的计算能力及CPU占用率,并根据所述计算单元的剩余计算能力分配计算量,其中,所述剩余计算能力为计算单元的总计算能力与其CPU占用率的差值;a calculation unit total controller, configured to record a calculation capability and a CPU occupancy rate of the plurality of calculation units, and allocate a calculation amount according to a remaining calculation capability of the calculation unit, wherein the remaining calculation capability is a total calculation capability of the calculation unit and The difference in CPU usage;
计算单元,设置为对业务板卡、网管控制单元、磁盘阵列中的数据进行计算处理,并通过计算单元总控制器实时上报计算单元的处理结果。The calculation unit is configured to perform calculation processing on the data in the service board, the network management unit, and the disk array, and report the processing result of the calculation unit in real time through the total controller of the calculation unit.
具体地,如图4所示,为本发明实施例中数据计算的流程图,在本发明实施例中,数据处理单元能够对业务板卡、网管控制单元、磁盘阵列中的数据进行处理,采用了分布式计算方法,能够将整个机架的所有计算资源进行独立,大大提高了机架计算资源的利用率。与图5中现有技术数据计算的流程相比,本发明实施例中的计算单元总控制器能够记录多个计算单元的计算能力及CPU占用率,而现有技术中,每个单独板卡使用独立的CPU和内存,因而避免了现有技术中部分板卡已经饱和而其他板卡却处于空闲的状况,且能够根据用户需求增减数据处理办卡的数量,既节约了成本,又便于后续的控制与维护。另外,本发明实施例中的数据处理单元还支持热插拔,且插拔过程中数据不会丢失,增加了机架的可维护性和机架的使用寿命。Specifically, as shown in FIG. 4, which is a flowchart of data calculation in the embodiment of the present invention, in the embodiment of the present invention, the data processing unit can process data in the service board, the network management unit, and the disk array, and adopt The distributed computing method can separate all computing resources of the entire rack, greatly improving the utilization of rack computing resources. Compared with the flow of the prior art data calculation in FIG. 5, the computing unit overall controller in the embodiment of the present invention can record the computing power and CPU usage of multiple computing units, and in the prior art, each individual board The use of independent CPU and memory avoids the fact that some boards in the prior art are saturated while other boards are idle, and the number of data processing cards can be increased or decreased according to user requirements, which saves cost and is convenient. Subsequent control and maintenance. In addition, the data processing unit in the embodiment of the present invention also supports hot plugging, and data is not lost during the plugging and unplugging process, which increases the maintainability of the rack and the service life of the rack.
磁盘阵列,设置为存储网管控制单元中的数据及各业务板卡的状态信息以及业务板卡的日志信息。The disk array is configured to store data in the network management unit and status information of each service board and log information of the service board.
具体地,如图6所示,为本发明实施例中日志管理的流程图,在本发明实施例中,不仅能够业务板卡的日志信息直接存放于磁盘阵列中,而且业务板卡还能够直接更新和读取磁盘阵列中的数据,与图7中现有技术日志管理流程相比,避免了更新和读取操作要先经过控制板卡再到网管服务器再到数据库的通讯方式,大大提升了数据传输的传递效率,并保证了数 据传输的正确性。另外,磁盘阵列也能够支持热插拔,且插拔过程中数据不会丢失。Specifically, as shown in FIG. 6 , which is a flowchart of the log management in the embodiment of the present invention, in the embodiment of the present invention, not only the log information of the service card can be directly stored in the disk array, but also the service board can directly Update and read the data in the disk array, compared with the prior art log management process in Figure 7, avoiding the update and read operations, first through the control board and then to the network management server to the database communication mode, greatly improved Data transfer efficiency and guaranteed number According to the correctness of the transmission. In addition, the disk array can also support hot swapping, and data will not be lost during the plugging process.
进一步地,在本发明实施例中,磁盘阵列还可以作为上述在网管控制单元上运行的网络管理软件的数据库,设置为存储网络管理软件的相关数据以及网管控制单元中的数据。Further, in the embodiment of the present invention, the disk array can also be used as the database of the network management software running on the network management unit, and is configured to store related data of the network management software and data in the network management unit.
在本发明实施例中,业务板卡为将光信号转化为电信号的板卡,并设置为接收和发送光业务信号。其中,业务板卡设置于第一机架外壳内,并通过第一机架外壳上的背板接口与网关控制单元连接。In the embodiment of the present invention, the service card is a card that converts an optical signal into an electrical signal, and is configured to receive and transmit an optical service signal. The service board is disposed in the first rack enclosure and connected to the gateway control unit through a backplane interface on the first rack enclosure.
具体地,如图8所示,为一体化机架800的结构图,即网管控制单元、数据处理单元、磁盘阵列以及业务板卡均安装于同一机架上,这种配置方式更加便于机架的安装配置。另外,还可以根据实际情况还可以增设第二机架,第一机架外壳与第二机架外壳通过数据接口进行连接,且第二机架外壳内还设有多个计算单元和业务板卡,其中,所述多个计算单元和业务板卡通过第二机架外壳上的接口以及第一机架外壳上的接口与位于第一机架外壳内的网关控制单元连接。具体地,如图9所示,其中,主机架901与从机架903的结构图如图所示,主机架901与从机架903通过分别位于主从机架上的数据接口进行数据的传递,主机架上的计算单元设置为处理主机架上网管控制单元的数据计算,而从机架上的计算单元设置为业务板卡上的数据计算。这种主从机架配置的方式更适用于业务量处理较大的站点。其中,软件定义网络SDN网络构架的机架可作为从机架,能够便于SDN网络架构处理业务数据的控制面。Specifically, as shown in FIG. 8 , the structural diagram of the integrated rack 800, that is, the network management unit, the data processing unit, the disk array, and the service board are all installed on the same rack, which is more convenient for the rack. Installation configuration. In addition, a second rack may be added according to actual conditions. The first rack shell and the second rack shell are connected through a data interface, and the second rack shell is further provided with a plurality of computing units and service boards. The plurality of computing units and service boards are connected to the gateway control unit located in the first rack housing through an interface on the second rack housing and an interface on the first rack housing. Specifically, as shown in FIG. 9, wherein the main frame 901 and the slave chassis 903 have the structural diagrams as shown in the figure, the main frame 901 and the slave chassis 903 transmit data through the data interfaces respectively located on the master and slave racks. The computing unit on the mainframe is configured to process the data calculation of the mainframe web control unit, and the computing unit from the rack is set to calculate the data on the service board. This master-slave rack configuration is more suitable for sites with large traffic handling. The rack of the software-defined network SDN network architecture can be used as a slave rack, which can facilitate the control plane of the SDN network architecture to process service data.
实施例二Embodiment 2
本发明实施例中公开了一种通讯方法,如图10所示,该方法1000包括以下步骤:A communication method is disclosed in the embodiment of the present invention. As shown in FIG. 10, the method 1000 includes the following steps:
步骤S1001:获取用户指令,根据用户指令进行处理及格式转换;Step S1001: Acquire a user instruction, and perform processing and format conversion according to a user instruction;
步骤S1003:根据转换指令进行数据处理,获取处理结果及日志信息;Step S1003: Perform data processing according to the conversion instruction, and obtain processing results and log information;
其中,根据转换指令进行数据处理,获取处理结果及日志信息(步骤 S1003)具体包括:The data processing is performed according to the conversion instruction, and the processing result and the log information are acquired (step S1003) specifically includes:
获取所有计算单元的剩余计算能力,其中,剩余计算能力为计算单元的总计算能力与其中央处理器CPU占用率的差值;Obtaining the remaining computing power of all computing units, wherein the remaining computing power is the difference between the total computing power of the computing unit and its CPU CPU usage;
选择所有计算单元中剩余计算能力较强的计算单元对转换指令进行数据处理,获取处理结果及日志信息。The calculation unit with the remaining computing power remaining in all the computing units is selected to perform data processing on the conversion instruction, and the processing result and the log information are obtained.
步骤S1005:将处理结果及日志信息进行存储。Step S1005: Store the processing result and the log information.
具体地,在本发明实施例中,网管控制单元接收用户发送的指令,并对磁盘阵列中存储的数据处理单元的处理结果及各业务板卡的状态信息进行查询,然后将用户指令发送到相应的业务板卡进行处理,并向用户反馈执行结果。业务板卡处理完成后,实时将记录的日志发送到磁盘阵列中进行存储。数据处理单元对对网管控制单元以及业务板卡上需要计算的数据进行计算,其中,在进行数据计算时,要先通过计算单元总控制器记录当前所有计算单元的总计算能力和CPU占用率,并将计算单元总计算能力与CPU占用率之间的差值记为计算单元的剩余计算能力,然后选择剩余所有计算单元中计算能力较强的计算单元,优先进行数据计算。主要采用了分布式计方法,实现了计算资源的均衡分配,避免了计算资源无法共享的问题,大大提高了计算资源的利用率。Specifically, in the embodiment of the present invention, the network management unit receives the instruction sent by the user, queries the processing result of the data processing unit stored in the disk array, and the status information of each service card, and then sends the user instruction to the corresponding The business board is processed and the results are reported back to the user. After the service board is processed, the recorded logs are sent to the disk array for storage in real time. The data processing unit calculates the data that needs to be calculated on the network management unit and the service board. When performing data calculation, the total calculation capability and CPU usage of all current computing units are first recorded by the calculation unit total controller. The difference between the total computing power of the computing unit and the CPU usage is recorded as the remaining computing power of the computing unit, and then the computing unit with the strong computing power among all remaining computing units is selected, and the data calculation is prioritized. The distributed metering method is mainly used to realize the balanced allocation of computing resources, avoiding the problem that computing resources cannot be shared, and greatly improving the utilization of computing resources.
显然,本领域的技术人员应该明白,上述的本发明的各模块或各步骤可以用通用的计算装置来实现,它们可以集中在单个的计算装置上,或者分布在多个计算装置所组成的网络上,可选地,它们可以用计算装置可执行的程序代码来实现,从而,可以将它们存储在存储装置中由计算装置来执行,并且在某些情况下,可以以不同于此处的顺序执行所示出或描述的步骤,或者将它们分别制作成各个集成电路模块,或者将它们中的多个模块或步骤制作成单个集成电路模块来实现。这样,本发明不限制于任何特定的硬件和软件结合。It will be apparent to those skilled in the art that the various modules or steps of the present invention described above can be implemented by a general-purpose computing device that can be centralized on a single computing device or distributed across a network of multiple computing devices. Alternatively, they may be implemented by program code executable by the computing device such that they may be stored in the storage device by the computing device and, in some cases, may be different from the order herein. The steps shown or described are performed, or they are separately fabricated into individual integrated circuit modules, or a plurality of modules or steps thereof are fabricated as a single integrated circuit module. Thus, the invention is not limited to any specific combination of hardware and software.
以上所述仅为本发明的优选实施例而已,并不用于限制本发明,对于本领域的技术人员来说,本发明可以有各种更改和变化。凡在本发明的精 神和原则之内,所作的任何修改、等同替换、改进等,均应包含在本发明的保护范围之内。The above description is only the preferred embodiment of the present invention, and is not intended to limit the present invention, and various modifications and changes can be made to the present invention. Where in the essence of the invention Any modifications, equivalent substitutions, improvements, etc. made within the principles of God and the principles are intended to be included within the scope of the present invention.
工业实用性Industrial applicability
如上所述,本发明实施例提供的一种机架及通讯方法具有以下有益效果:实现了模块化管理和对用户命令并行下发,简化了机房安装配置,能够更好地满足业务发生变化时的需求。 As described above, the rack and the communication method provided by the embodiments of the present invention have the following beneficial effects: modular management and concurrent delivery of user commands, simplifying the installation and configuration of the equipment room, and better meeting the change of the service. Demand.

Claims (6)

  1. 一种机架,包括,机架外壳,还包括:A rack, including a rack housing, further comprising:
    设置于第一机架外壳内,并通过第一机架外壳上的背板接口与外部设备连接的网管控制单元、磁盘阵列以及数据处理单元,a network management unit, a disk array, and a data processing unit disposed in the first rack housing and connected to the external device through the backplane interface on the first rack housing,
    所述网管控制单元,设置为将用户指令进行处理及格式转换后发送给业务板卡、数据处理单元、磁盘阵列,并将业务板卡反馈的业务状态发送给用户;The network management unit is configured to process and format the user instructions, and then send the information to the service board, the data processing unit, and the disk array, and send the service status fed back by the service board to the user;
    所述数据处理单元,设置为对业务板卡、网管控制单元、磁盘阵列中的数据进行处理,并实时上报所述数据处理单元的处理结果;The data processing unit is configured to process data in the service board, the network management unit, and the disk array, and report the processing result of the data processing unit in real time;
    所述磁盘阵列,设置为存储所述网管控制单元中的数据及各业务板卡的状态信息以及业务板卡的日志信息。The disk array is configured to store data in the network management unit and status information of each service card and log information of the service board.
  2. 如权利要求1所述的机架,其中,所述数据处理单元包括:位于背板上的计算单元总控制器以及与所述计算单元总控器连接的多个计算单元,其中,The rack of claim 1 wherein said data processing unit comprises: a computing unit master controller located on the backplane and a plurality of computing units coupled to said computing unit master controller, wherein
    所述计算单元总控制器,设置为记录所述多个计算单元的计算能力及CPU占用率,并根据所述计算单元的剩余计算能力分配计算量,其中,所述剩余计算能力为计算单元的总计算能力与其CPU占用率的差值;The computing unit general controller is configured to record a computing capability and a CPU occupancy rate of the plurality of computing units, and allocate a calculation amount according to a remaining computing capability of the computing unit, wherein the remaining computing capability is a computing unit The difference between the total computing power and its CPU usage;
    所述计算单元,设置为对业务板卡、网管控制单元、磁盘阵列中的数据进行计算处理,并通过计算单元总控制器实时上报所述计算单元的处理结果。The calculation unit is configured to perform calculation processing on the data in the service board, the network management unit, and the disk array, and report the processing result of the calculation unit in real time through the calculation unit total controller.
  3. 如权利要求1所述的机架,其中,还包括:设置于第一机架外壳内,并通过第一机架外壳上的背板接口与网管控制单元连接的业务板卡;The rack of claim 1, further comprising: a service board disposed in the first rack housing and connected to the network management unit through a backplane interface on the first rack housing;
    所述业务板卡为将光信号转化为电信号的板卡,并设置为接收和发送光业务信号。The service card is a card that converts an optical signal into an electrical signal and is configured to receive and transmit optical service signals.
  4. 如权利要求1所述的机架,其中,还包括:The rack of claim 1 further comprising:
    与第一机架外壳连接的第二机架外壳,以及设置于所述第二机架外壳内的多个计算单元和多个业务板卡。a second rack housing coupled to the first rack housing, and a plurality of computing units and a plurality of service boards disposed within the second rack housing.
  5. 一种通讯方法,包括: A communication method comprising:
    获取用户指令,根据所述用户指令进行处理及格式转换;Obtaining a user instruction, performing processing and format conversion according to the user instruction;
    根据所述转换指令进行数据处理,获取处理结果及日志信息;Performing data processing according to the conversion instruction, and obtaining processing results and log information;
    将所述处理结果及日志信息进行存储。The processing result and log information are stored.
  6. 如权利要求5所述的方法,其中,根据所述转换指令进行数据处理,获取处理结果及日志信息的步骤包括:The method of claim 5, wherein the step of performing data processing according to the conversion instruction, obtaining the processing result and the log information comprises:
    获取所有计算单元的剩余计算能力,其中,所述剩余计算能力为计算单元的总计算能力与其中央处理器CPU占用率的差值;Obtaining a remaining computing power of all computing units, wherein the remaining computing power is a difference between a total computing power of the computing unit and a CPU usage of the central processing unit;
    选择所有计算单元中所述剩余计算能力较强的计算单元对所述转换指令进行数据处理,获取处理结果及日志信息。 The computing unit with the remaining computing power in all the computing units is selected to perform data processing on the conversion instruction, and the processing result and the log information are obtained.
PCT/CN2015/089191 2014-12-05 2015-09-08 Rack and communication method WO2016086700A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201410736306.3A CN105721956A (en) 2014-12-05 2014-12-05 Frame and communication method
CN201410736306.3 2014-12-05

Publications (1)

Publication Number Publication Date
WO2016086700A1 true WO2016086700A1 (en) 2016-06-09

Family

ID=56090963

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2015/089191 WO2016086700A1 (en) 2014-12-05 2015-09-08 Rack and communication method

Country Status (2)

Country Link
CN (1) CN105721956A (en)
WO (1) WO2016086700A1 (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113810313A (en) * 2021-09-06 2021-12-17 山石网科通信技术股份有限公司 Processing method and processing device for distributed session message

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108415357A (en) * 2018-02-09 2018-08-17 上海欣诺通信技术股份有限公司 A kind of optical transmission device effectively reducing power consumption
CN111614430A (en) * 2020-05-13 2020-09-01 深圳纤亿通科技有限公司 Active wavelength division equipment with online solution function

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN2606495Y (en) * 2003-01-16 2004-03-10 深圳市飞奔网络技术有限公司 Optical fibre transceiver with internal netting tube
US20050031347A1 (en) * 2003-07-03 2005-02-10 Soto Alexander I. Communication system and method for an optical local area network
CN2788461Y (en) * 2004-11-15 2006-06-14 深圳市飞通宽带技术有限公司 Remote end fiber transmitting-receiving device capable of detecting optical communication link circuit at any moment
CN1825788A (en) * 2006-03-24 2006-08-30 上海全光网络科技股份有限公司 Light path automatic switching system
CN101604261A (en) * 2009-07-08 2009-12-16 深圳先进技术研究院 The method for scheduling task of supercomputer
CN101719791A (en) * 2009-11-24 2010-06-02 江苏科技大学 Optic fiber transceiver having remote monitoring function and data transmission method
CN101834663A (en) * 2010-04-08 2010-09-15 中兴通讯股份有限公司 Detection method and system of optical fiber connection

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN2606495Y (en) * 2003-01-16 2004-03-10 深圳市飞奔网络技术有限公司 Optical fibre transceiver with internal netting tube
US20050031347A1 (en) * 2003-07-03 2005-02-10 Soto Alexander I. Communication system and method for an optical local area network
CN2788461Y (en) * 2004-11-15 2006-06-14 深圳市飞通宽带技术有限公司 Remote end fiber transmitting-receiving device capable of detecting optical communication link circuit at any moment
CN1825788A (en) * 2006-03-24 2006-08-30 上海全光网络科技股份有限公司 Light path automatic switching system
CN101604261A (en) * 2009-07-08 2009-12-16 深圳先进技术研究院 The method for scheduling task of supercomputer
CN101719791A (en) * 2009-11-24 2010-06-02 江苏科技大学 Optic fiber transceiver having remote monitoring function and data transmission method
CN101834663A (en) * 2010-04-08 2010-09-15 中兴通讯股份有限公司 Detection method and system of optical fiber connection

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113810313A (en) * 2021-09-06 2021-12-17 山石网科通信技术股份有限公司 Processing method and processing device for distributed session message
CN113810313B (en) * 2021-09-06 2024-04-19 山石网科通信技术股份有限公司 Processing method and processing device for distributed session message

Also Published As

Publication number Publication date
CN105721956A (en) 2016-06-29

Similar Documents

Publication Publication Date Title
CN102546813B (en) A kind of High-Performance Computing Cluster computing system based on x86 PC framework
CN101594235B (en) Method for managing blade server based on SMBUS
US9609782B2 (en) Rack assembly structure
CN106155959A (en) Data transmission method and data transmission system
CN103986786A (en) Remote cloud desktop operation system
CN100562850C (en) Multi-processor load distribution-regulation method
CN103955441B (en) Equipment management system, equipment management method and IO (Input/Output) expansion interface
CN103533081A (en) Cloud-computing-based billing system and implementation method thereof
KR102126440B1 (en) Power distribution to mobile devices
US20120297100A1 (en) Storage system and data transmission method
CN211628241U (en) PCIE bus structure switching topology through software
WO2022037265A1 (en) Edge computing center integrated server
WO2016086700A1 (en) Rack and communication method
CN105472291A (en) Digital video recorder with multiprocessor cluster and realization method of digital video recorder
CN105099776A (en) Cloud server management system
CN105162623A (en) Cloud processing method and cloud server
CN113099410B (en) 5G power edge data transmission processing method, device, terminal and medium
US11055252B1 (en) Modular hardware acceleration device
CN102799708B (en) Graphic processing unit (GPU) high-performance calculation platform device applied to electromagnetic simulation
US20180205784A1 (en) Fiber-Based Distributed Data Center Architecture
CN109977053A (en) The network-based shared I/O interface design method of one kind and system
CN212302372U (en) Edge computing center integrated server
CN110309031B (en) Load balancing micro-computing cluster architecture
CN106294062A (en) A kind of method by computing function in server with management function separate design
CN201417410Y (en) Computer motherboard

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 15864980

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 15864980

Country of ref document: EP

Kind code of ref document: A1