WO2018059052A1 - 一种数据处理系统和波分设备 - Google Patents

一种数据处理系统和波分设备 Download PDF

Info

Publication number
WO2018059052A1
WO2018059052A1 PCT/CN2017/091437 CN2017091437W WO2018059052A1 WO 2018059052 A1 WO2018059052 A1 WO 2018059052A1 CN 2017091437 W CN2017091437 W CN 2017091437W WO 2018059052 A1 WO2018059052 A1 WO 2018059052A1
Authority
WO
WIPO (PCT)
Prior art keywords
ethernet service
service data
data
wavelength division
mapping
Prior art date
Application number
PCT/CN2017/091437
Other languages
English (en)
French (fr)
Inventor
朱治宇
Original Assignee
华为技术有限公司
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 华为技术有限公司 filed Critical 华为技术有限公司
Publication of WO2018059052A1 publication Critical patent/WO2018059052A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00—Arrangements for detecting or preventing errors in the information received
    • H04L1/0001—Systems modifying transmission characteristics according to link quality, e.g. power backoff
    • H04L1/0002—Systems modifying transmission characteristics according to link quality, e.g. power backoff by adapting the transmission rate
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04J—MULTIPLEX COMMUNICATION
    • H04J14/00—Optical multiplex systems
    • H04J14/02—Wavelength-division multiplex systems
    • H04J14/0201—Add-and-drop multiplexing
    • H04J14/0215—Architecture aspects
    • H04J14/022—For interconnection of WDM optical networks
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/50—Network services
    • H04L67/56—Provisioning of proxy services
    • H04L67/565—Conversion or adaptation of application format or content
    • H04L67/5651—Reducing the amount or size of exchanged application data

Definitions

  • the present invention relates to the field of network communication technologies, and in particular, to an Ethernet service data processing system and a wavelength division device applied in a data center interconnection (DCI) application scenario.
  • DCI data center interconnection
  • the data center provides a high-end data transmission service and high-speed access service, which provides professional services for enterprise Internet access and enterprise IT management, enabling enterprises and individuals to quickly conduct business through the network.
  • Data center interconnection refers to network connection and data interaction between different data centers, and realizes point-to-point interconnection between different data centers.
  • the application scenario of the data center interconnection conforms to the Muxponder feature of the Optical Transmission Network (OTN) device.
  • OTN Optical Transmission Network
  • the mapping method defined by the ITU-T G.709 protocol is adopted.
  • all different size granules such as 10GE, 25GE, 40GE, 50GE, 100GE, and 400GE, must be sent to different sizes before being sent.
  • the granular Ethernet service data is sequentially aggregated and encapsulated into a large bandwidth ODU (Optical Data Unit) container, and the Ethernet service data can be transmitted through the ODU container processing.
  • ODU Optical Data Unit
  • the asynchronous mapping method for Ethernet service data is currently also mapped by the ITU-T G.709 protocol.
  • the data center needs to encapsulate the Ethernet service data of all rates into the ODU container corresponding to the rate, so as to implement the scheduling function of the Ethernet service data.
  • the specific mapping path is shown in Figure 1. among them:
  • Ethernet service data For 40GE Ethernet service data, it is first subjected to TTT (Timing Transparent Transcoding) transcoding compression processing, and then GMP mapping is used to map it to the ODU3 container for the Ethernet service in the ODU3 container. The data continues to be mapped to the ODUCn container using GMP mapping.
  • TTT Transmission Transparent Transcoding
  • the BMP mapping mode is first used to map it to the ODU4 container. Then, the Ethernet service data in the ODU4 container is continuously mapped to the ODUCn container by using the GMP mapping mode.
  • Ethernet service data For 25GE, 50GE, or 400GE Ethernet service data, first map it by BMP mapping. Go to the ODUflex container, and then continue to adopt the GMP mapping method for the Ethernet service data in the ODUflex container and map it to the ODUCn container.
  • the basic processing idea of the ITU-T G.709 protocol is to first map the Ethernet service data into a low-order ODUk container or a low-order ODUflex container (this mapping step is called a first-level mapping), and k is equal to 2e. 3 or 4, and then map the Ethernet service data in the low-order ODUk container or the low-order ODUflex container to the high-order ODUCn container (this mapping step is called the second-level mapping), that is, the ITU-T G.709 protocol.
  • a two-level mapping method is defined.
  • the inventor of the present application has found that in the application scenario of data center interconnection, different data centers are separated by means of wavelength division equipment (such as optical fiber), and there is no small particle ODU layer (ie, low-order ODUk container or low-order ODUflex).
  • wavelength division equipment such as optical fiber
  • ODU layer ie, low-order ODUk container or low-order ODUflex.
  • the scheduling and management requirements of the container so if the two-level mapping method defined by the ITU-T G.709 protocol is used to implement point-to-point interconnection between different data centers, the first level mapping step is redundant.
  • the application scenario of the data center interconnection is sensitive to absolute delay, and the two-level mapping method defined by the ITU-T G.709 protocol obviously needs to go through two mapping processes, and the absolute delay is large.
  • the present application provides a data processing system and a wavelength division device to simplify the mapping process of Ethernet service data and reduce the absolute delay.
  • a first aspect of the present application provides a data processing system including a plurality of data centers, each of which includes a wavelength division device, and each of the data centers is interconnected based on the wavelength division device, the plurality of data
  • the center includes at least a first data center and a second data center.
  • the wavelength division device of the first data center uses the universal mapping procedure GMP mapping mode to map the Ethernet service data to the optical data unit ODUCn container.
  • the first data center in the present application directly maps the Ethernet service data to be transmitted into the ODUCn container by using the GMP mapping manner.
  • the present application maps the two levels. Reduced to a first-level mapping, simplifies the mapping process of Ethernet service data, and reduces the absolute delay.
  • the Ethernet service data to be transmitted is Ethernet service data of any rate.
  • the Ethernet service data of any rate is mapped to the ODUCn container by using the GMP mapping mode, which simplifies the mapping process of the Ethernet service data, reduces the absolute delay, and reduces the design complexity.
  • the method further includes: after receiving the target network service data sent by the first data center, the wavelength division device of the second data center adopts a universal mapping
  • the GMP mapping method is used to demap the target Ethernet service data to obtain Ethernet service data.
  • the second data center after receiving the target Ethernet service data sent by the first data center, the second data center demaps the target Ethernet service data to obtain the Ethernet.
  • Network service data Generally, the Ethernet service data obtained by the demapping is small-particle Ethernet service data, and the GMP demapping mode and the GMP mapping mode are mutually reversed.
  • the second data center obtains small-particle Ethernet service data by adopting GMP de-mapping method, and the present application decodes two levels compared to the two-level demapping method defined by the ITU-T G.709 protocol.
  • the shot reduction is a first-level demapping, which simplifies the demapping process of the Ethernet service data and reduces the absolute delay.
  • the second data center is further configured to: after receiving the target Ethernet service data sent by the first data center, The network service data is allocated a rate, and the allocated rate is adjusted according to the difference between the received service rate of the target Ethernet service data and the rate allocated for the target Ethernet service data.
  • This application is based on the DCI application scenario, and there is no need for accurate data rate synchronization. Therefore, it is not necessary to set multiple PLLs (Phase Locked Loop) to restore the rate of Ethernet service data, and only need to receive the Ethernet locally.
  • the service data is allocated a rate and adjusted accordingly. This application further simplifies the demapping process of the Ethernet service data and reduces the design complexity.
  • a second aspect of the present application provides a wavelength division device, including: a determining module, configured to determine Ethernet service data to be transmitted; and a mapping module, configured to use the universal mapping procedure GMP mapping manner, to use the Ethernet service data It is mapped into the optical data unit ODUCn container to obtain the target Ethernet service data.
  • the Ethernet service data to be transmitted is Ethernet service data of any rate.
  • the method further includes: a receiving module, configured to receive the target Ethernet service data; and a demapping module, configured to adopt a universal mapping procedure GMP mapping method, Decoding the target Ethernet service data to obtain Ethernet service data.
  • the method further includes: an allocating module, configured to allocate a rate for the target Ethernet service data; and an adjusting module, configured to: according to the received target The rate is adjusted by the difference between the service rate of the network service data and the rate at which the allocation module allocates the target Ethernet service data.
  • Figure 1 is a schematic diagram of a mapping path defined by an existing ITU-T G.709 protocol
  • FIG. 2 is a schematic structural diagram of a network system in which data centers are interconnected in an embodiment of the present application
  • FIG. 3 is a schematic structural diagram of interconnection between data centers in an embodiment of the present application.
  • FIG. 4 is a schematic structural diagram of a wavelength division device according to an embodiment of the present application.
  • FIG. 5 is a flowchart of a data processing method according to an embodiment of the present application.
  • FIG. 6 is a schematic diagram of a mapping path disclosed in an embodiment of the present application.
  • FIG. 7 is a flowchart of another data processing method according to an embodiment of the present application.
  • FIG. 8 is a schematic structural diagram of a data processing apparatus according to an embodiment of the present disclosure.
  • FIG. 9 is a schematic structural diagram of another data processing apparatus according to an embodiment of the present disclosure.
  • FIG. 11 is a structural block diagram of a whole device in the embodiment of the present application.
  • the network system architecture of the data center interconnection includes multiple data center DCs, each of which includes a wavelength division device, and each data center DC is interconnected based on a wavelength division device, as shown in FIG.
  • data transmission is realized by using a fiber to extend the transmission distance.
  • the structure of the wavelength division device is as shown in FIG. 4, and includes a wavelength division board and a combiner/demultiplexer.
  • the wavelength division board includes a QSFP+/QSFP28 (Quad Small Form-factor Pluggable), an oDSP (Optical Digital Signal Processing) chip, and a CFP coherent optical module.
  • the QSFP+/QSFP28 is a short-range optical module that interfaces with routers or storage arrays in the data center to access Ethernet service data of different sizes, such as 10GE, 25GE, 40GE, 50GE, 100GE, and 400GE.
  • the oDSP chip implements mapping or demapping processing of the Ethernet service data
  • the CFP coherent optical module is a long-distance dense wavelength division multiplexing optical module, which is used for docking with the WDM device in the data center in an off-site.
  • mapping or demapping the Ethernet service data implemented by the oDSP chip in the present invention will be described in detail by the inventors.
  • each data center includes uplink data transmission and downlink data transmission.
  • the uplink data transmission specifically refers to the data center that asynchronously maps and aggregates the Ethernet service data of different sizes into a large bandwidth ODUCn container, processes it through the ODUCn container, and sends it to the remote data center.
  • Data transmission specifically refers to the data center receiving Ethernet service data sent from an off-site data center, and de-mapping the received Ethernet service data to obtain Ethernet service data of various sizes of particles, and thus obtaining The process of sending Ethernet service data of various sizes and sizes to the router.
  • FIG. 5 is a flowchart of a data processing method provided by the present application.
  • the first data center is used as the sending end of the Ethernet service data.
  • the data center serves as the receiving end of the Ethernet service data, and the method specifically includes:
  • Step 101 The first data center determines Ethernet service data to be transmitted.
  • the Ethernet service data may be Ethernet service data of any rate.
  • Step 102 The wavelength division device of the first data center adopts a GMP mapping manner, and maps the Ethernet service data into an ODUCn container to obtain target Ethernet service data.
  • Step 103 The wavelength division device of the first data center sends the target Ethernet service data to the wavelength division device of the second data center.
  • the mapping principle of step 102 in the embodiment of the present application and the mapping principle of the Ethernet service data in the low-order ODUk container or the low-order ODUflex container defined in the ITU-T G.709 protocol are mapped to the high-order ODUCn container.
  • the target Ethernet service data in this application refers to large-particle Ethernet service data.
  • the description of the large-particle Ethernet service data and the small-particle Ethernet service data in this application is a relative concept.
  • This application obtains the target Ethernet service data after the original Ethernet service data is processed through GMP mapping, because the original Ethernet service data.
  • the granularity is smaller than the granularity of the target Ethernet service data. Therefore, the original Ethernet service data is referred to as small-particle Ethernet service data, and the obtained target Ethernet service data is referred to as large-particle Ethernet service data.
  • the ODUCn container in the present application may allocate corresponding partial time slots ODUCn.ts for Ethernet service data of different size particles in advance, and specifically, allocate a part of time slots ODUCn for 10GE Ethernet service data.
  • Ts the part of the time slot ODUCn.ts is recorded as ODUCn.ts1
  • a part of the time slot ODUCn.ts is allocated for the 40GE Ethernet service data
  • the part of the time slot ODUCn.ts is recorded as ODUCn.ts2, which is 100GE Ethernet service data.
  • a part of the time slot ODUCn.ts is allocated, and the part of the time slot ODUCn.ts is recorded as ODUCn.ts3, and a part of the time slot ODUCn.ts is allocated for the Ethernet service data of 25GE, 50GE and 400GE, and the part of the time slot ODUCn.ts is recorded as ODUCn. .ts4.
  • ODUCn.ts1, ODUCn.ts2, ODUCn.ts3, and ODUCn.ts4 are spliced together to be a large bandwidth ODUCn.
  • each data center is interconnected by a direct connection of a wavelength division device, and there is no need for scheduling and management of ODUs at different levels. Therefore, the application will be 10GE, 25GE, 40GE, 50GE, 100GE,
  • the 400GE Ethernet service data is directly mapped to the ODUCn container by using GMP mapping. Because the present application directly maps the Ethernet service data to be transmitted into the ODUCn container by using the GMP mapping manner, the present application reduces the two-level mapping to the primary mapping manner compared to the two-level mapping manner defined by the ITU-T G.709 protocol. It simplifies the mapping process of Ethernet service data, and at the same time reduces the absolute delay, and the delay of Ethernet service data can be reduced by up to 1 us.
  • FIG. 7 is another flowchart of a data processing method provided by the present application, and specifically includes:
  • Step 201 The wavelength division device of the second data center receives the target Ethernet service data sent by the first data center.
  • Step 202 The wavelength division device of the second data center uses GMP de-mapping to demap the target Ethernet service data to obtain Ethernet service data.
  • the Ethernet service data obtained by the demapping is small-particle Ethernet service data.
  • the second data center in the embodiment of the present application serves as a receiving end for receiving Ethernet service data.
  • the second data center uses the GMP mapping method to set the target Ethernet service data.
  • De-mapping obtains relatively small-particle Ethernet service data, where GMP demapping mode and GMP mapping mode are mutually reversed. Therefore, in the present application, the second data center obtains small-particle Ethernet service data by adopting GMP de-mapping method, and the present application reduces the two-level demapping manner compared to the two-level demapping method defined by the ITU-T G.709 protocol. It is a one-level demapping, which simplifies the demapping process of Ethernet service data and reduces the absolute delay.
  • the embodiment of the present application may further include:
  • Step 203 The second data center allocates a rate to the target Ethernet service data.
  • Step 204 The second data center adjusts the allocated rate according to the difference between the received service rate of the target Ethernet service data and the rate allocated for the target Ethernet service data.
  • the DCI application scenario does not require the precise synchronization of the data rate. Therefore, it is not necessary to set a plurality of PLLs (Phase Locked Loops) to restore the rate of the Ethernet service data.
  • PLLs Phase Locked Loops
  • a self-vibrating clock is locally set.
  • a rate is allocated for the received target Ethernet service data.
  • the self-oscillation clock complies with the standard service clock frequency offset range defined in the 802.3 protocol.
  • step 205 of the present application can be implemented as follows:
  • Step 2051 Calculate the difference between the service rate of the received target Ethernet service data and the service rate allocated locally for the target Ethernet service data.
  • Step 2051 generally obtains the difference by means of an asynchronous FIFO (First Input First Output) watermark and clock equalization means.
  • asynchronous FIFO First Input First Output
  • Step 2052 Parse the idle codeword in the service flow packet gap of the target Ethernet service data.
  • Step 2053 when the rate of the received target Ethernet service data is faster than the rate allocated locally for the target Ethernet service data, the idle codeword is deleted under an appropriate opportunity; when the rate ratio of the received target Ethernet service data is received The local rate of allocation of the target Ethernet service data is slow, and the idle codeword is inserted at an appropriate opportunity.
  • the present application implements the adjustment of the rate value by adding or deleting the Idle codeword by internally parsing the Idle (idle codeword) codeword.
  • multiple PLLs are not required to be set compared to the ITU-T G.709 protocol, which further simplifies the demapping process of the Ethernet service data and reduces the design complexity.
  • the present application further discloses a wavelength division device.
  • the structure of the wavelength division device is as shown in FIG. 8, and includes: a determination module 10, a mapping module 20, and a transmission module 30.
  • the determining module 10, the mapping module 20, and the sending module 30 are working, the data processing method shown in FIG. 5 is executed, as follows:
  • the determining module 10 is configured to determine Ethernet service data to be transmitted; wherein the Ethernet service data is Ethernet service data of any rate.
  • the mapping module 20 is configured to map the Ethernet service data into the ODUCn container by using a GMP mapping manner to obtain target Ethernet service data.
  • the sending module 30 is configured to send the target Ethernet service data.
  • the wavelength division device in the embodiment of the present application determines the Ethernet service data to be transmitted through the determining module 10, and uses the mapping module 20 to map the Ethernet service data to the ODUCn container to obtain the target Ethernet service.
  • the data is further transmitted by the transmitting module 30 to the target Ethernet service data. Therefore, the wavelength division device in the embodiment of the present application directly maps the Ethernet service data to be transmitted into the ODUCn container by using a GMP mapping manner, compared to the ITU-
  • This application reduces the two-level mapping to the first-level mapping Shooting simplifies the mapping process of Ethernet service data and reduces the absolute delay.
  • the wavelength division device provided by the present application further includes a receiving module 40 and a demapping module 50, as shown in FIG.
  • the receiving module 40 and the demapping module 50 are working, the data processing method shown in FIG. 7 is executed, as follows:
  • the receiving module 40 is configured to receive target Ethernet service data.
  • the demapping module 50 is configured to demap the target Ethernet service data to obtain Ethernet service data by using a GMP demapping method.
  • the data processing apparatus in this embodiment obtains Ethernet service data (ie, small-particle Ethernet service data) by using GMP de-mapping method, and the present application compares the two-level demapping method defined by the ITU-T G.709 protocol.
  • the two-level demapping is reduced to one-level demapping, which simplifies the demapping process of Ethernet service data and reduces the absolute delay.
  • the wavelength division device further comprises: an allocation module 60 and an adjustment module 70. among them,
  • An allocating module 60 configured to allocate a rate for the target Ethernet service data
  • the adjusting module 500 is configured to adjust the rate according to the difference between the received service rate of the target Ethernet service data and the rate allocated by the allocating module 60 for the target Ethernet service data.
  • the present application also provides an oDSP chip capable of implementing the data processing method shown in FIG. 5 and FIG. 7 above.
  • FIG. 10 is a structural block diagram of the original device
  • FIG. 11 is a structural block diagram of the whole device in the present application.
  • the mapping and demapping functions for the Ethernet service data cannot be integrated into the oDSP chip. Therefore, Framer+ is used on the original board.
  • the design structure of the oDSP chip, and the mapping and demapping process of the Ethernet service data is greatly simplified after the application is integrated into the oDSP chip, so the present application is on the original board.
  • the design structure of a single oDSP chip is adopted. The application reduces the overall power consumption, cost, and overall integration of the system.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Quality & Reliability (AREA)
  • Optical Communication System (AREA)
  • Small-Scale Networks (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

本申请提供一种数据处理系统和波分设备，系统包括多个数据中心，每个数据中心均包括一波分设备，各数据中心之间基于所述波分设备实现互联，所述多个数据中心至少包括第一数据中心和第二数据中心。其中，第一数据中心确定待传输的以太网业务数据后，由第一数据中心的波分设备采用GMP映射方式，将所述以太网业务数据映射到ODUCn容器中，得到目标以太网业务数据，并将所述目标以太网业务数据发送至第二数据中心的波分设备。本申请相比于ITU-T G.709协议定义的两级映射方式，本申请将两级映射减少为一级映射，简化了以太网业务数据的映射处理过程，且减小了绝对延时。

Description

一种数据处理系统和波分设备
本申请要求于2016年09月29日提交中国专利局、申请号为201610867950.3、发明名称为“一种数据处理系统和波分设备”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及网络通信技术领域，尤其涉及一种应用于数据中心互联(Data Center Interconnection，简称DCI)应用场景中的以太网业务数据处理系统和波分设备。
背景技术
数据中心提供了一种高端的数据传输服务和高速接入服务，其为企业上网、企业IT管理提供专业服务，使得企业和个人能够迅速借助网络开展业务。数据中心互联则指多个不同的数据中心之间进行网络连接、数据交互，实现了不同数据中心之间点对点互联。
数据中心互联的应用场景特性符合光传送网(Optical Transmission Network，简称OTN)设备的Muxponder(复用器)特性。在OTN网络中，采用ITU-T G.709协议定义的映射方式。基于OTN网络具有的对所有层面的ODU的调度和管理需求，所有不同大小颗粒，如10GE、25GE、40GE、50GE、100GE、400GE的以太网业务数据在被发送前，需要先将各种不同大小颗粒的以太网业务数据依次经过异步映射、封装汇聚到一个大带宽的ODU(Optical Data Unit,光学数据单元)容器中，通过ODU容器处理才能实现以太网业务数据的发送。
在目前的数据中心互联的应用场景中，对于以太网业务数据的异步映射方法，目前也是采用ITU-T G.709协议定义的映射方式。数据中心需要将所有速率的以太网业务数据分别封装到与之对应速率的ODU容器中，才能实现对以太网业务数据的调度功能。具体的映射路径如图1所示。其中：
1、对于10GE的以太网业务数据，首先采用BMP(Bit-synchronous Mapping Procedure，比特同步映射规程)映射方式，将其映射到ODU2e容器中，进而对于ODU2e容器中的以太网业务数据则继续采用GMP(Generic Mapping Procedure，通用映射规程)映射方式，将其映射到ODUCn容器中。
2、对于40GE的以太网业务数据，首先经过TTT(Timing Transparent Transcoding，定时透明转码)转码压缩处理，进而采用GMP映射方式，将其映射到ODU3容器中，对于ODU3容器中的以太网业务数据则继续采用GMP映射方式，将其映射到ODUCn容器中。
3、对于100GE的以太网业务数据，首先采用BMP映射方式，将其映射到ODU4容器中，进而对于ODU4容器中的以太网业务数据则继续采用GMP映射方式，将其映射到ODUCn容器中。
4、对于25GE、50GE或400GE的以太网业务数据，首先采用BMP映射方式，将其映射 到ODUflex容器中，进而对于ODUflex容器中的以太网业务数据则继续采用GMP映射方式，将其映射到ODUCn容器中。
因此，ITU-T G.709协议的基本处理思想是先将以太网业务数据映射到低阶ODUk容器中或低阶ODUflex容器中(该映射步骤称之为第一级映射)，k等于2e、3或4，再将低阶ODUk容器中或低阶ODUflex容器中的以太网业务数据映射到高阶ODUCn容器中(该映射步骤称之为第二级映射)，即ITU-T G.709协议定义了两级映射方式。
本申请的发明人发现，在数据中心互联的应用场景下，不同数据中心之间借助于波分设备(如光纤)拉远传输距离，不存在小颗粒ODU层面(即低阶ODUk容器或低阶ODUflex容器)的调度和管理需求，因此如果采用ITU-T G.709协议定义的两级映射方式来实现不同数据中心之间点对点互联，则第一级映射步骤冗余。且数据中心互联的应用场景对于绝对延时敏感，而采用ITU-T G.709协议定义的两级映射方式，显然需要经历两次映射过程，绝对延时大。
发明内容
有鉴于此，本申请提供一种数据处理系统和波分设备，以简化以太网业务数据的映射处理过程，减小绝对延时。
本申请的第一方面提供了一种数据处理系统，包括多个数据中心，每个数据中心均包括一波分设备，各数据中心之间基于所述波分设备实现互联，所述多个数据中心至少包括第一数据中心和第二数据中心。其中，第一数据中心确定待传输的以太网业务数据后，由所述第一数据中心的波分设备采用通用映射规程GMP映射方式，将所述以太网业务数据映射到光学数据单元ODUCn容器中，得到目标以太网业务数据，并将所述目标以太网业务数据发送至所述第二数据中心的波分设备。本申请中的第一数据中心直接将待传输的以太网业务数据采用GMP映射方式映射到ODUCn容器中，相比于ITU-T G.709协议定义的两级映射方式，本申请将两级映射减少为一级映射，简化了以太网业务数据的映射处理过程，且减小了绝对延时。
结合第一方面，在第一方面的第一种实现方式中，所述待传输的以太网业务数据为任意速率的以太网业务数据。本申请针对任意速率的以太网业务数据均使用GMP映射方式映射到ODUCn容器中，在简化了以太网业务数据的映射处理过程，减小绝对延的同时，降低了设计复杂度。
结合第一方面，在第一方面的第二种实现方式中，还包括：所述第二数据中心的波分设备接收到所述第一数据中心发送的目标太网业务数据后，采用通用映射规程GMP解映射方式，将所述目标太网业务数据解映射得到以太网业务数据。对于本申请接收侧的第二数据中心来说，第二数据中心接收到第一数据中心发送的目标以太网业务数据后，采用GMP解映射方式，将所述目标以太网业务数据解映射得到以太网业务数据，一般的，该解映射得到的以太网业务数据为小颗粒以太网业务数据，其中GMP解映射方式与GMP映射方式互逆。本申请中，第二数据中心通过采用GMP解映射方式得到小颗粒以太网业务数据，相比于ITU-T G.709协议定义的两级解映射方式，本申请将两级解映 射减少为一级解映射，简化了以太网业务数据的解映射处理过程，且减小了绝对延时。
结合第一方面，在第一方面的第三种实现方式中，所述第二数据中心还用于，在接收到所述第一数据中心发送的目标以太网业务数据后，为所述目标以太网业务数据分配一速率，并根据接收到的所述目标以太网业务数据的业务速率与为所述目标以太网业务数据分配的速率之差，对所述分配的速率进行调整。
本申请基于DCI应用场景不存在数据速率精准同步的需求，因此不需要设置多个PLL(Phase Locked Loop，锁相环)来恢复以太网业务数据的速率，只需在本地为接收到的以太网业务数据分配一速率，并相应对其进行调整即可，本申请进一步简化了以太网业务数据的解映射处理过程，降低了设计复杂度，
本申请的第二方面提供了一种波分设备，包括：确定模块，用于确定待传输的以太网业务数据；映射模块，用于采用通用映射规程GMP映射方式，将所述以太网业务数据映射到光学数据单元ODUCn容器中，得到目标以太网业务数据。
结合第二方面，在第二方面的第一种实现方式中，所述待传输的以太网业务数据为任意速率的以太网业务数据。
结合第二方面，在第二方面的第二种实现方式中，还包括：接收模块，用于接收所述目标以太网业务数据；解映射模块，用于采用通用映射规程GMP解映射方式，将所述目标以太网业务数据解映射得到以太网业务数据。
结合第二方面，在第二方面的第三种实现方式中，还包括：分配模块，用于为所述目标以太网业务数据分配一速率；调整模块，用于根据接收到的所述目标以太网业务数据的业务速率与所述分配模块为所述目标以太网业务数据分配的速率之差，对所述速率进行调整。
附图说明
为了更清楚地说明本申请实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请的实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据提供的附图获得其他的附图。
图1为现有ITU-T G.709协议定义的映射路径示意图；
图2为本申请实施例中数据中心互联的网络系统架构示意图；
图3为本申请实施例中数据中心之间互联的结构示意图；
图4为本申请实施例中波分设备的结构示意图；
图5为本申请实施例公开的一种数据处理方法的流程图；
图6为本申请实施例公开的一种映射路径示意图；
图7为本申请实施例公开的另一种数据处理方法的流程图；
图8为本申请实施例公开的一种数据处理装置的结构示意图；
图9为本申请实施例公开的另一种数据处理装置的结构示意图；
图10为现有技术中整机设备的结构框图；
图11为本申请实施例中整机设备的结构框图。
具体实施方式
下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本申请保护的范围。
如图2所示，数据中心互联的网络系统架构包括多个数据中心DC，每个数据中心DC均包括一波分设备，各个数据中心DC之间基于波分设备实现互联，如图3所示，如借助于光纤拉远传输距离实现数据传输。
具体的，波分设备的结构如图4所示，包括波分单板和合波器/分波器。其中，波分单板上包括QSFP+/QSFP28(Quad Small Form-factor Pluggable，四通道SFP接口)、oDSP(Optical Digital Signal Process，光数字信号处理)芯片、CFP相干光模块。QSFP+/QSFP28为短距光模块，用于与数据中心中的路由器或存储阵列对接，从而实现接入不同大小颗粒，如10GE、25GE、40GE、50GE、100GE、400GE的以太网业务数据。oDSP芯片实现对以太网业务数据的映射或解映射处理，CFP相干光模块为长距密集波分复用光模块，用于实现与异地的数据中心中的波分设备对接。
对于本发明中oDSP芯片实现的对以太网业务数据的映射或解映射处理的实现过程，下面发明人将进行详细说明。
首先发明人需要说明的是，每个数据中心都包括上行数据传输和下行数据传输。其中上行数据传输具体指的是数据中心将各种不同大小颗粒的以太网业务数据经过异步映射、封装汇聚到一个大带宽的ODUCn容器中，通过ODUCn容器处理后发送至异地数据中心的过程，下行数据传输具体指的是数据中心接收到来自于异地数据中心发送的以太网业务数据，并对接收到的以太网业务数据解映射后得到各种不同大小颗粒的以太网业务数据，进而将得到的各种不同大小颗粒的以太网业务数据发送至路由器的过程。
对于上行数据传输的过程如图5所示，其示出了本申请提供的一种数据处理方法的流程图，在本申请实施例中，第一数据中心作为以太网业务数据的发送端，第二数据中心作为以太网业务数据的接收端，方法具体包括：
步骤101，第一数据中心确定待传输的以太网业务数据。
其中，所述以太网业务数据可以为任意速率的以太网业务数据。
步骤102，第一数据中心的波分设备采用GMP映射方式，将所述以太网业务数据映射到ODUCn容器中，得到目标以太网业务数据。
步骤103，第一数据中心的波分设备将所述目标以太网业务数据发送至第二数据中心的波分设备。
其中本申请实施例步骤102的映射原理与ITU-T G.709协议中定义的将低阶ODUk容器中或低阶ODUflex容器中的以太网业务数据映射到高阶ODUCn容器中的映射原理一 致。本申请中目标以太网业务数据指的是大颗粒以太网业务数据。本申请中的大颗粒以太网业务数据与小颗粒以太网业务数据的描述是一个相对概念，本申请将原以太网业务数据经过GMP映射处理后得到目标以太网业务数据，由于原以太网业务数据的颗粒度小于目标以太网业务数据的颗粒度，所以本申请将原以太网业务数据称之为小颗粒以太网业务数据，得到的目标以太网业务数据称之为大颗粒以太网业务数据。
结合图6所示，本申请中的ODUCn容器可以预先为不同大小颗粒的以太网业务数据分配相应的部分时隙ODUCn.ts，具体的，可以为10GE的以太网业务数据分配一部分时隙ODUCn.ts，该部分时隙ODUCn.ts记为ODUCn.ts1，为40GE的以太网业务数据分配一部分时隙ODUCn.ts，该部分时隙ODUCn.ts记为ODUCn.ts2，为100GE的以太网业务数据分配一部分时隙ODUCn.ts，该部分时隙ODUCn.ts记为ODUCn.ts3，为25GE、50GE以及400GE的以太网业务数据分配一部分时隙ODUCn.ts，该部分时隙ODUCn.ts记为ODUCn.ts4。ODUCn.ts1、ODUCn.ts2、ODUCn.ts3和ODUCn.ts4拼接起来就是大带宽的ODUCn。
在本申请实施例中，各数据中心之间通过波分设备的直连实现互联，并不存在对不同层面的ODU的调度和管理需求，因此本申请将10GE、25GE、40GE、50GE、100GE、400GE的以太网业务数据均采用GMP映射方式直接映射到ODUCn容器中。因为本申请直接将待传输的以太网业务数据采用GMP映射方式映射到ODUCn容器中，相比于ITU-T G.709协议定义的两级映射方式，本申请将两级映射减少为一级映射，简化了以太网业务数据的映射处理过程，且同时减小了绝对延时，以太网业务数据的延时最多可减小达1us。
对于下行数据传输的过程如图7所示，其示出了本申请提供的一种数据处理方法的另一种流程图，具体包括：
步骤201，第二数据中心的波分设备接收第一数据中心发送的目标以太网业务数据。
步骤202，第二数据中心的波分设备采用GMP解映射方式，将所述目标以太网业务数据解映射得到以太网业务数据。
这里，解映射得到的以太网业务数据为小颗粒以太网业务数据。
本申请实施例中的第二数据中心作为接收以太网业务数据的接收端。结合图6所示，对于接收侧的第二数据中心来说，第二数据中心接收到第一数据中心发送的目标以太网业务数据后，采用GMP解映射方式，将所述目标以太网业务数据解映射得到相对的小颗粒以太网业务数据，其中GMP解映射方式与GMP映射方式互逆。因此在本申请中，第二数据中心通过采用GMP解映射方式得到小颗粒以太网业务数据，相比于ITU-T G.709协议定义的两级解映射方式，本申请将两级解映射减少为一级解映射，简化了以太网业务数据的解映射处理过程，且减小了绝对延时。
此外本申请实施例在步骤201后，还可以包括：
步骤203，第二数据中心为所述目标以太网业务数据分配一速率。
步骤204，第二数据中心根据接收到的所述目标以太网业务数据的业务速率与为所述目标以太网业务数据分配的速率之差，对所述分配的速率进行调整。
ITU-T G.709协议中涉及的下行数据传输过程中，以太网业务数据的速率需要透传， 因此在下行数据传输过程中需要设置多个PLL(Phase Locked Loop，锁相环)用以恢复各以太网业务数据的速率。
而本申请中，DCI应用场景不存在数据速率精准同步的需求，因此不需要设置多个PLL(Phase Locked Loop，锁相环)来恢复以太网业务数据的速率。本申请中各数据中心本地设置有自振时钟，当作为接收端的第二数据中心接收到目标以太网业务数据后，便为接收到的目标以太网业务数据分配一速率。其中，自振时钟符合802.3协议中定义的标准业务时钟频偏范围。
在本申请实施例中，自振时钟为目标以太网业务数据分配的速率可能存在偏差，因此，本申请会根据接收到的目标以太网业务数据的业务速率与为目标以太网业务数据分配的速率之差，对所述分配的速率进行调整。具体的，本申请步骤205可以采用如下方式实现：
步骤2051，计算接收到的目标以太网业务数据的业务速率与本地为目标以太网业务数据分配的业务速率之差。
步骤2051一般通过异步FIFO(First Input First Output，先入先出队列)水线、时钟鉴相等手段得到该差值。
步骤2052，解析目标以太网业务数据的业务流包间隙中的空闲码字。
步骤2053，当接收到的目标以太网业务数据的速率比本地为目标以太网业务数据分配的速率快，则在适当的机会下删除空闲码字；当接收到的目标以太网业务数据的速率比本地为目标以太网业务数据分配的速率慢，则在适当的机会下插入空闲码字。
因此，本申请通过芯片内部解析Idle(空闲码字)码字，通过增加或删除Idle码字来实现对速率值大小的调整。
因此本申请实施例公开的下行数据传输过程中，相比于ITU-T G.709协议不需要设置多个PLL，进一步简化了以太网业务数据的解映射处理过程，降低了设计复杂度，
基于前述实施例公开的数据处理系统和方法，本申请还公开了一种波分设备，该波分设备的结构如图8所示，包括：确定模块10、映射模块20和发送模块30。当确定模块10、映射模块20和发送模块30工作时，执行图5所示的数据处理方法，具体如下：
确定模块10，用于确定待传输的以太网业务数据；其中所述以太网业务数据为任意速率的以太网业务数据。
映射模块20，用于采用GMP映射方式，将所述以太网业务数据映射到ODUCn容器中，得到目标以太网业务数据。
发送模块30，用于发送所述目标以太网业务数据。
本申请实施例中的波分设备通过确定模块10确定待传输的以太网业务数据，并利用映射模块20采用GMP映射方式，将所述以太网业务数据映射到ODUCn容器中，得到目标以太网业务数据，进而利用发送模块30发送所述目标以太网业务数据，因此本申请实施例中的波分设备直接将待传输的以太网业务数据采用GMP映射方式映射到ODUCn容器中，相比于ITU-T G.709协议定义的两级映射方式，本申请将两级映射减少为一级映 射，简化了以太网业务数据的映射处理过程，且减小了绝对延时。
此外本申请提供的波分设备还包括，接收模块40和解映射模块50，如图9所示。当接收模块40和解映射模块50工作时，执行图7所示的数据处理方法，具体如下：
接收模块40，用于接收目标以太网业务数据；
解映射模块50，用于采用GMP解映射方式，将所述目标以太网业务数据解映射得到以太网业务数据。
本实施例中的数据处理装置通过采用GMP解映射方式得到以太网业务数据(即小颗粒以太网业务数据)，相比于ITU-T G.709协议定义的两级解映射方式，本申请将两级解映射减少为一级解映射，简化了以太网业务数据的解映射处理过程，且减小了绝对延时。
作为更优的，波分设备还可以进一步包括：分配模块60和调整模块70。其中，
分配模块60，用于为所述目标以太网业务数据分配一速率；
调整模块500，用于根据接收到的所述目标以太网业务数据的业务速率与所述分配模块60为所述目标以太网业务数据分配的速率之差，对所述速率进行调整。
此外本申请还提供了一种oDSP芯片，该oDSP芯片能够实现前述图5、图7所示的数据处理方法。
结合图10、图11所示，图10为原整机设备的结构框图，图11为本申请中整机设备的结构框图。安装按照ITU-T G.709协议的标准进行设计时，由于oDSP芯片面积、功耗问题，对于以太网业务数据的映射、解映射功能无法集成到oDSP芯片，故而在原有单板上采用Framer+oDSP芯片的设计结构，而本申请对于以太网业务数据的映射、解映射处理过程大大简化后，对于以太网业务数据的映射、解映射功能能够集成到oDSP芯片内部，因此本申请在原有单板上采用单oDSP芯片的设计结构。本申请降低了系统整体功耗、成本、提高了整机集成度。
需要说明的是，本说明书中的各个实施例均采用递进的方式描述，每个实施例重点说明的都是与其他实施例的不同之处，各个实施例之间相同相似的部分互相参见即可。对于系统实施例而言，由于其与方法实施例基本相似，所以描述的比较简单，相关之处参见方法实施例的部分说明即可。
最后，还需要说明的是，在本文中，诸如第一和第二等之类的关系术语仅仅用来将一个实体或者操作与另一个实体或操作区分开来，而不一定要求或者暗示这些实体或操作之间存在任何这种实际的关系或者顺序。而且，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者设备不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者设备所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括所述要素的过程、方法、物品或者设备中还存在另外的相同要素。
以上对本申请所提供的数据处理系统和波分设备进行了详细介绍，本文中应用了具体个例对本申请的原理及实施方式进行了阐述，以上实施例的说明只是用于帮助理解本申请的方法及其核心思想；同时，对于本领域的一般技术人员，依据本申请的思想，在具体实施方式及应用范围上均会有改变之处，综上所述，本说明书内容不应理解为对本申请的限制。

Claims (8)

  1. 一种数据处理系统，包括多个数据中心，每个数据中心均包括一波分设备，各数据中心之间基于所述波分设备实现互联，所述多个数据中心至少包括第一数据中心和第二数据中心，其特征在于，
    所述第一数据中心确定待传输的以太网业务数据后，由所述第一数据中心的波分设备采用通用映射规程GMP映射方式，将所述以太网业务数据映射到光学数据单元ODUCn容器中，得到目标以太网业务数据，并将所述目标以太网业务数据发送至所述第二数据中心的波分设备。
  2. 根据权利要求1所述的系统，其特征在于，所述待传输的以太网业务数据为任意速率的以太网业务数据。
  3. 根据权利要求1或2所述的系统，其特征在于，还包括：
    所述第二数据中心的波分设备接收到所述第一数据中心发送的目标太网业务数据后，采用通用映射规程GMP解映射方式，将所述目标太网业务数据解映射得到以太网业务数据。
  4. 根据权利要求3所述的系统，其特征在于，所述第二数据中心还用于，在接收到所述第一数据中心发送的目标以太网业务数据后，为所述目标以太网业务数据分配一速率，并根据接收到的所述目标以太网业务数据的业务速率与为所述目标以太网业务数据分配的速率之差，对所述分配的速率进行调整。
  5. 一种波分设备，其特征在于，包括：
    确定模块，用于确定待传输的以太网业务数据；
    映射模块，用于采用通用映射规程GMP映射方式，将所述以太网业务数据映射到光学数据单元ODUCn容器中，得到目标以太网业务数据；
    发送模块，用于发送所述目标以太网业务数据。
  6. 根据权利要求5所述的波分设备，其特征在于，所述待传输的以太网业务数据为任意速率的以太网业务数据。
  7. 根据权利要求5或6所述的波分设备，其特征在于，还包括：
    接收模块，用于接收所述目标以太网业务数据；
    解映射模块，用于采用通用映射规程GMP解映射方式，将所述目标以太网业务数据解映射得到以太网业务数据。
  8. 根据权利要求7所述的波分设备，其特征在于，还包括：
    分配模块，用于为所述目标以太网业务数据分配一速率；
    调整模块，用于根据接收到的所述目标以太网业务数据的业务速率与所述分配模块为所述目标以太网业务数据分配的速率之差，对所述速率进行调整。
PCT/CN2017/091437 2016-09-29 2017-07-03 一种数据处理系统和波分设备 WO2018059052A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201610867950.3 2016-09-29
CN201610867950.3A CN106549732B (zh) 2016-09-29 2016-09-29 一种数据处理系统和波分设备

Publications (1)

Publication Number Publication Date
WO2018059052A1 true WO2018059052A1 (zh) 2018-04-05

Family

ID=58368305

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/091437 WO2018059052A1 (zh) 2016-09-29 2017-07-03 一种数据处理系统和波分设备

Country Status (2)

Country Link
CN (1) CN106549732B (zh)
WO (1) WO2018059052A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116232983A (zh) * 2022-12-29 2023-06-06 中国联合网络通信集团有限公司 一种以太网业务的路由确定方法、装置、设备及存储介质

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106549732B (zh) * 2016-09-29 2019-11-29 华为技术有限公司 一种数据处理系统和波分设备

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101902275A (zh) * 2010-08-20 2010-12-01 中兴通讯股份有限公司 监控信息的发送/接收方法、装置及传递系统
CN103825668A (zh) * 2009-12-24 2014-05-28 华为技术有限公司 通用映射规程gmp映射方法、解映射方法及装置
CN104247452A (zh) * 2012-04-26 2014-12-24 中兴通讯股份有限公司 一种光通道数据单元的gmp映射方法及装置
CN106549732A (zh) * 2016-09-29 2017-03-29 华为技术有限公司 一种数据处理系统和波分设备

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101834688B (zh) * 2009-03-09 2011-08-31 华为技术有限公司 光传送网中的映射、解映射方法及装置
CN102377482B (zh) * 2010-08-26 2016-08-03 中兴通讯股份有限公司 一种光纤通道业务故障传递方法及装置

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103825668A (zh) * 2009-12-24 2014-05-28 华为技术有限公司 通用映射规程gmp映射方法、解映射方法及装置
CN101902275A (zh) * 2010-08-20 2010-12-01 中兴通讯股份有限公司 监控信息的发送/接收方法、装置及传递系统
CN104247452A (zh) * 2012-04-26 2014-12-24 中兴通讯股份有限公司 一种光通道数据单元的gmp映射方法及装置
CN106549732A (zh) * 2016-09-29 2017-03-29 华为技术有限公司 一种数据处理系统和波分设备

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116232983A (zh) * 2022-12-29 2023-06-06 中国联合网络通信集团有限公司 一种以太网业务的路由确定方法、装置、设备及存储介质
CN116232983B (zh) * 2022-12-29 2024-05-03 中国联合网络通信集团有限公司 一种以太网业务的路由确定方法、装置、设备及存储介质

Also Published As

Publication number Publication date
CN106549732A (zh) 2017-03-29
CN106549732B (zh) 2019-11-29

Similar Documents

Publication Publication Date Title
US9225462B2 (en) Method, apparatus and system for transmitting and receiving client signals
JP4878629B2 (ja) 多重伝送システムおよび多重伝送方法
US7873073B2 (en) Method and system for synchronous high speed Ethernet GFP mapping over an optical transport network
US9497064B2 (en) Method and apparatus for transporting ultra-high-speed Ethernet service
US10218455B2 (en) Method and apparatus for increasing and decreasing variable optical channel bandwidth
US11177931B2 (en) Clock transmission method and related device
WO2010121520A1 (zh) 光传送网的信号传送方法、设备及通信系统
US20120163812A1 (en) Method and apparatus for transmitting packet data over optical transport network
JP5313351B2 (ja) １０ギガビット光ファイバーチャネルサービスを光伝送ネットワークに伝送する方法及び装置
WO2010130076A1 (zh) 光传输网中的数据传送方法、系统及设备
WO2018059052A1 (zh) 一种数据处理系统和波分设备
WO2018090296A1 (zh) 无源光网络的通信方法、装置和系统
JP5736964B2 (ja) 伝送装置及びデータ伝送方法
WO2019214582A1 (zh) 业务信号传输方法及装置
JP2011041218A (ja) 伝送装置、伝送方法および多重回路
WO2010135864A1 (zh) 传送客户数据的方法、设备及通信系统
KR100628330B1 (ko) 스위칭 기능을 가진 광 트랜스폰더
JP4695171B2 (ja) 光ディジタル送信装置、光ディジタル送受信システムおよび光ディジタル送信方法
JP5856661B1 (ja) フレームデータ処理方法
JP5856662B1 (ja) フレーム多重方法
EP2784956B1 (en) Asymmetric OTN network traffic support
KR100927599B1 (ko) 종속신호를 통합하는 방법 및 통합 접속보드
CN118074850A (zh) 传输数据的方法及装置
EP2988436A1 (en) Method to operate a network unit and network unit
JP2012195736A (ja) 送受信装置および送受信方法

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: 17854506

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: 17854506

Country of ref document: EP

Kind code of ref document: A1