WO2005018154A1 - A method supporting the multilevel schedule of multi-port and multi-traffic - Google Patents

A method supporting the multilevel schedule of multi-port and multi-traffic Download PDF

Info

Publication number
WO2005018154A1
WO2005018154A1 PCT/CN2004/000466 CN2004000466W WO2005018154A1 WO 2005018154 A1 WO2005018154 A1 WO 2005018154A1 CN 2004000466 W CN2004000466 W CN 2004000466W WO 2005018154 A1 WO2005018154 A1 WO 2005018154A1
Authority
WO
WIPO (PCT)
Prior art keywords
service
level
user
scheduling
services
Prior art date
Legal status (The legal status 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 status listed.)
Ceased
Application number
PCT/CN2004/000466
Other languages
English (en)
French (fr)
Inventor
Jie Yu
Zhengang Li
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
ZTE Corp
Original Assignee
ZTE Corp
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 ZTE Corp filed Critical ZTE Corp
Publication of WO2005018154A1 publication Critical patent/WO2005018154A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/50—Queue scheduling
    • H04L47/62—Queue scheduling characterised by scheduling criteria
    • H04L47/6215—Individual queue per QOS, rate or priority
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L12/00—Data switching networks
    • H04L12/28—Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
    • H04L12/42—Loop networks
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/10—Flow control; Congestion control
    • H04L47/24—Traffic characterised by specific attributes, e.g. priority or QoS
    • H04L47/2441—Traffic characterised by specific attributes, e.g. priority or QoS relying on flow classification, e.g. using integrated services [IntServ]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/50—Queue scheduling
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/50—Queue scheduling
    • H04L47/56—Queue scheduling implementing delay-aware scheduling
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/50—Queue scheduling
    • H04L47/60—Queue scheduling implementing hierarchical scheduling

Definitions

  • Multi-level scheduling method supporting multiple ports and multiple services
  • the present invention relates to a data frame communication technology, and in particular, to an elastic packet ring of a metropolitan area network.
  • RPR Remote Packet Ring
  • Resilient packet ring technology is an emerging technology, its purpose is to build data frames with reusable bandwidth, each node has a fair algorithm to ensure bandwidth occupancy, and has ring protection and Q0S (Quality of Service) capabilities.
  • Transmission networks are mainly targeted at metro backbone ring networks and metro access ring networks.
  • the IEEE specifically established the IEEE802.17 standard group, and formulated the IEEE802.17 MAC (Media Access Control) layer standard based on the flexible packet ring technology that can form flexible packet rings.
  • Type A is a real-time service
  • Type B is divided into two parts.
  • B-CIR committed informat ion rate
  • B-EIR extraction information rate
  • class C is a best effort service.
  • the three types of services are distinguished by the Service Class defined in the frame structure. The services of a single elastic packet ring site entering the elastic packet ring loop are respectively shaped and scheduled according to the three types of A, B, and C, and then enter a certain elastic packet ring loop.
  • an elastic packet ring device generally includes two elastic packet ring loop-side ports and a set of user-side ports.
  • the frames entered and output from the elastic packet ring loop-side port are elastic packet ring MAC frames, with A , B, and C service classification identifiers, and most of the user-side ports are Ethernet ports. Frames entering and outputting from this port are Ethernet frames, and the frame structure does not carry eight, B, and C service classification identifiers.
  • Multiple elastic packet ring devices form an elastic packet ring by using the ports on the ring side of the elastic packet ring.
  • Existing Flexible Packet Ring There is no queue on the user-side port in the device.
  • a data frame After a data frame enters a user-side port, it enters the A / B / C queue of a flexible packet ring loop after a certain forwarding operation, and then performs standard integer scheduling. Operate, enter the queue to be sent, and finally in a certain flexible packet ring loop, as shown in FIG. 2.
  • the above-mentioned method for processing and processing frames received by a user-side port has some problems: First, there are often multiple user-side ports on an elastic packet ring device, and a certain port may have a large amount of traffic and a certain port may have a small amount of traffic.
  • the service entered from a user-side port may be a type A service, a type B service, or a type C service. If a large number of type C services are entered first, and then a type A service is entered, the service is based on first come Principle, it will cause a large number of low-priority type C services to be processed preferentially, while high-priority type A services can only be processed after type C services, which will not solve the delay and delay jitter problems of type A services. .
  • the technical problem to be solved by the present invention is to provide a multi-level scheduling method that supports multi-port and multi-service.
  • the elastic packet ring device there are multiple user-side ports, and the existing ports have unreasonable service scheduling defects. Meet the requirements of the elastic packet ring.
  • the multi-stage scheduling method supporting multi-port and multi-service includes the following steps: A user-side port of a single elastic packet ring device is configured with a secondary queue of service levels A, B, and C, and the user-side port The received data frame is stored in the corresponding second-level queue according to the service classification identifier; the second-level scheduling of the user-side port's second-level queue is first-level service-level scheduling and then port-weighted scheduling;
  • the data frames that have undergone the two-level scheduling enter the first-level queues of the A, B, and C service levels, respectively; and the three types of services in the first-level queue are subjected to flexible packet ring-on-loop scheduling.
  • the minimum amount of A, B, and C services that each user-side port allows to enter the elastic packet ring loop The bandwidth is configured by the user.
  • the service level scheduling is based on the priority order of Class A services over Class B services, and Class B services over Class C services.
  • the port weighted scheduling is based on the weights of various services at each user-side port.
  • the weight value is obtained according to the ratio of the minimum bandwidth of the three types of services of the user-side ports A, B, and C to the bandwidth of the similar services of all the user-side ports of the flexible packet ring device.
  • the weights of the different types of service levels may be different.
  • the step of performing two-level scheduling on the second-level queue specifically includes:
  • Step 1) In turn, check whether the second-level queue of the A service level of the user-side port is empty. If it is empty, go to step 2); if it is not empty, go to step 3);
  • Step 2) Determine whether the user-side port is the last user-side port. If so, go to step 5); if not, go back to step 1) to query the second-level queue of the A service level of the next user-side port;
  • Step 3 According to the weight of the type A service of the user-side port, the type A service in the level 2 queue is dispatched to the level 1 queue of the level A service class;
  • Step 4) Determine whether the first-level queue of the class A service level is full, if not, go to step 2); If it is full, go to step 5);
  • Step 5 In turn, check whether the second-level queue of the B service level of the user-side port is empty. If it is empty, go to step 6); if it is not empty, go to step 7);
  • Step 6) Determine whether the user-side port is the last user-side port. If YES, go to step 9); if not, go back to step 5) to query the second-level queue of the B service level of the next user-side port ;
  • Step 7) Adjust the type B services in the secondary queue according to the weight of the type B services of the user-side port.
  • Step 8) determine whether the first-level queue of the class B service level is full, if not, go to step 6); if it is full, go to step 9);
  • Step 9) Query whether the second-level queue of the C service level of the user-side port is empty, and if it is empty, go to step 10); if it is not empty, go to step 11);
  • Step 10) Determine whether the user-side port is the last user-side port, and if yes, go to step 1); if not, return to step 9), and query the second-level queue of the C service level of the next user-side port;
  • Step 11 Based on the weight of the C-type service on the user-side port, the C-type service in the second-level queue is dispatched to the first-level queue of the C-level service class;
  • Step 12 Determine whether the first-level queue of the class C service level is full. If it is not full, go to step 10); if it is full, go to step 1).
  • the step of performing ring scheduling on the three types of services in the first-level queue further includes: shaping the three types of services separately, and then scheduling various types of service data frames to output to the corresponding elastic packet ring loop.
  • the shaping and scheduling are performed in accordance with the order of A type services priority B type services, B type services priority C type services.
  • the shaping is to shape the corresponding services in the first-level queue according to the credit value of the various services.
  • the credit value of the various types of services is determined by the total bandwidth of the A, B, and C services allowed by the elastic packet ring device to enter the elastic packet ring loop.
  • the total bandwidth of the A, B, and C services allowed by the single elastic packet ring device to enter the elastic packet ring loop is configured by the user.
  • the steps of scheduling on the ring specifically include:
  • Step 1) The credit value of the class A service is used to shape the class A service in the first-level queue and consume The corresponding credit value;
  • Step 2 Scheduling the shaped A service data frame and outputting it to the flexible packet ring;
  • Step 3 Determine whether the credit value of the A service is exhausted, and if it is not exhausted, return to step 1); if exhausted , Then suspend the shaping and scheduling of Class A services, and perform step 4);
  • Step 4) Shaping the B-type services in the first-level queue according to the credit value of the B-type services and consuming the corresponding credit value;
  • Step 5) Scheduling the shaped B service data frame and outputting it to the flexible packet ring;
  • Step 6) Determine whether the credit value of the B service is exhausted, and if it is not exhausted, return to step 4); if it has been exhausted , Then suspend the shaping and scheduling of Class B services, and perform step 7);
  • Step 7) Shaping the C services in the first-level queue according to the credit value of the C services and consuming the corresponding credit value;
  • Step 8) Scheduling the shaped C service data frame and outputting it to the flexible packet ring;
  • Step 9) Determine whether the credit value of the C service is exhausted, and if it is not exhausted, return to step 7); if it has been consumed When it is completed, the shaping and scheduling of the C-type service are suspended, and the process returns to step 1).
  • a second-level queue of three service levels of A, B, and C is configured on each user-side port, and a corresponding two-level scheduling mechanism of first-level service level scheduling and port weighted scheduling is provided for the second-level queue, It can meet the priority order of the first scheduling of Class A services, the second of Class B services, and the last scheduling of Class C services entered by each port, thereby ensuring the delay and delay jitter requirements of Class A services and Class B services. It also prevents starvation of each port in the same level of service.
  • the method of the present invention effectively solves the scheduling policy in the case where services with multiple user ports need to enter the elastic packet ring loop, and each user port has three types of services, B, and C at the same time.
  • FIG. 1 is a schematic diagram of an elastic packet ring device
  • FIG. 2 is a schematic diagram of a service scheduling framework in the prior art
  • FIG. 1 is a schematic diagram of an elastic packet ring device
  • FIG. 2 is a schematic diagram of a service scheduling framework in the prior art
  • FIG. 1 is a schematic diagram of an elastic packet ring device
  • FIG. 2 is a schematic diagram of a service scheduling framework in the prior art
  • FIG. 1 is a schematic diagram of an elastic packet ring device
  • FIG. 2 is a schematic diagram of a service scheduling framework in the prior art
  • FIG. 3 is a schematic framework diagram of a multi-level scheduling method according to the present invention.
  • FIG. 6 is a flowchart of performing a loop scheduling in the present invention.
  • FIG. 1 and FIG. 2 are introductions to the prior art, which have been described in detail in the background section above, and will not be repeated here.
  • each user-side port of the elastic packet ring device is configured with three queues, which serve as the second-level queues of the three service classes of A, B, and C.
  • the data frame received by the user-side port first enters the second-level queue of the corresponding service level, and performs two-level scheduling of service-level scheduling and port-weighted scheduling for each type of service data frame in the second-level queue.
  • the data frame then enters the first-level queues of the A, B, and C service levels, and then is reshaped by the shaper, and then sent to the queue to be sent, and finally enters the flexible packet ring loop.
  • a single elastic packet ring device is allowed to enter the total bandwidth of class A services, total bandwidth of class B services, and total bandwidth of class C services, and each user-side port is allowed to enter A of the elastic packet ring ring.
  • the minimum bandwidth of the class B service, the minimum bandwidth of the class B service, and the minimum bandwidth of the class C service can be configured by the user, where the sum of the minimum bandwidth of the class A service allowed by each user-side port and the total of the class A service of the flexible packet ring device The bandwidth is equal, and the above-mentioned correspondence relationship also exists for the type B service and the type C service.
  • Each user-side port receives a data frame, and then classifies the received data frame according to its classification identifier A, B, and C service levels, and simultaneously classifies the data. Frames are respectively placed in the second-level queues of the corresponding service class to which the port belongs.
  • the two-stage scheduling process is shown in Figure 5.
  • ⁇ ⁇ A single flexible packet ring site has N user-side ports.
  • First-level service level scheduling is performed first, that is, the type A service is scheduled first, then the type B service is scheduled, and finally the type C service is scheduled.
  • the service second-level queue otherwise, the service data frames in the second-level queue of the A-type service of the user-side port are dispatched to the first-level queue of the A-type service according to the weight of the type A service of the user-side port.
  • the weight value of the Class A service is obtained based on the ratio of the Class A service bandwidth of the user-side port to the Class A service bandwidth of the entire elastic packet ring device. If all the second-level queues of the class A service level are empty, or the first-level queues of the class A service level are full, the second-level queues of the class B service are scheduled.
  • the scheduling method of the Class B service level second queue is the same as the scheduling method of the Class A service level second queue described above. If the second-level queues of the class B service class of all ports are empty or the first-level queues of the class B service class are full, then the scheduling of the second class queue of the second class of service is performed, and the second class queue of the second class of service is scheduled. Similar to the scheduling method of the above-mentioned class A service level two queues, the above steps are repeated over and over again. It should be noted that the port weights of the three service classes of A, B, and C may not be ⁇ ! Same value.
  • Figure 6 shows the flowchart of the ring scheduling.
  • the normal elastic packet ring ring scheduling is performed for the services in the first-level queue, including the priority of the first class A service class, then the B class service class, and the last class C service class.
  • Hierarchical order The shaping operation and scheduling process of business data.
  • A-type services are divided into AO-type and A1-type.
  • the AO-type services are of a higher level. Therefore, when scheduling A-type services, they are scheduled in the order of AO-type services and then A1-type services.
  • the A-type services of the first-level queue are first shaped according to the credit value of the AO, and the corresponding AO credit value is consumed.
  • the shaped data frames are sent to the queue to be sent, and enter the flexible packet ring.
  • the scheduling of Al services is performed to shape the A services of the first-level queue according to the credit value of A1 and consume the corresponding A1 credit value.
  • the shaped data frame then passes through the D shaper and consumes the corresponding The credit value of D, the shaped data is sent to the queue to be sent, and enters the elastic packet ring.
  • the scheduling of Class A services is suspended, and then the scheduling of Class B services, and then the scheduling of Class C services in turn.
  • the process of scheduling Class B and C services is similar to that of Class A1 services.
  • the credit value of different types of services is determined by the total bandwidth of A, B, and C services that a single elastic packet ring device is allowed to enter into the elastic packet ring loop.

Landscapes

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

Description

一种支持多端口多业务的多级调度方法 技术领域
本发明涉及数据帧通信技术， 具体地说， 涉及城域网的弹性分组环
( Res i l ient Packet Ring, 简称 RPR )技术， 尤其涉及 RPR设备具有多个用户 側端口， 且每个用户端口支持多种业务的调度方法。
背景技术
弹性分组环技术是一项新兴技术， 其目的在于组建具有带宽可复用的、 各 个节点具有公平算法保证带宽占用率、 具有环保护和 Q0S (服务质量， Qual i ty of Service )能力的数据帧传输网络， 主要针对城域骨干环网和城域接入环网。 2000年 12月 IEEE专门成立了 IEEE802. 17标准组， 制定基于弹性分组环技术 的能够组建弹性分组环的 IEEE802. 17 MAC ( Media Access Control )层标准。
弹性分组环技术最主要的特征是对业务进行了分类， 所有的业务被分成 A、 B、 C三类： A类为实时业务； B类分成两个部分， B-CIR ( commi tted informat ion rate ) 为承诺速率等级业务， B-EIR ( excess informat ion rate ) 为超过承诺 速率等级的业务， C 类为尽力而为业务， 以上三类业务通过帧结构中定义的 Service Class来区分。单个弹性分组环站点进入弹性分组环环路的业务根据 A、 B、 C三类分别进行整形和调度， 之后进入某个弹性分组环环路中。
如图 1所示， 弹性分组环设备一般包括 2个弹性分组环环路侧端口和一组 用户侧端口， 从弹性分组环环路侧端口进入和输出的帧是弹性分组环 MAC帧 , 具有 A、 B、 C业务分类标识， 而用户侧端口多数是以太网端口， 从该端口进入 和输出的帧是以太网帧， 其帧结构中不带八、 B、 C业务分类标识。 多个弹性分 组环设备通过弹性分组环环路侧端口组成一个弹性分组环。 现有的弹性分组环 设备中用户侧端口是没有队 的， 当数据帧进入某个用户侧端口后， 经过一定 的转发操作后进入某个弹性分组环环路的 A/B/C队列， 然后进行标准的整型调 度操作， 进入待发送队列中， 最后 某个弹性分组环环路中， 如图 2所示。 上述关于用户侧端口接收的帧的调度处理方法存在一些问题： 首先， 弹性分组 环设备的用户侧端口往往有多个， 可能出现某个端口业务量大， 而某个端口业 务量小的情况， 现有技术一般是采用先来先服务的方式进行调度， 结果会导致 业务量小的端口出现饥饿现象。 其次， 从某个用户侧端口进入的业务可能是 A 类业务或 B类业务或 C类业务， 如果先进入的是大量的 C类业务， 而后进入的 是 A类业务， 那么根据先来先服务的原则， 会导致大量的低优先级的 C类业务 被优先处理， 而高优先级的 A类业务只能在 C类业务以后处理， 这样就无法解 决 A类业务的时延和时延抖动问题。
发明内容
本发明所要解决的技术问题是提供一种支持多端口多业务的多级调度方 法， 针对弹性分组环设备存在多个用户側端口， 而现有端口间业务调度不合理 的缺陷， 以更好地满足弹性分组环的要求。
本发明所述的支持多端口多业务的多级调度方法， 包括以下步骤： 在单个弹性分组环设备的每个用户侧端口配置 A、 B、 C业务等级的二级队 列，所述用户侧端口接收的数据帧根据业务分类标识存放在相应的二级队列中； 对所述用户侧端口的二级队列进行先业务等级调度后端口加权调度的两级 调度；
经过两级调度的数据帧分别进入 A、 B、 C业务等级的一级队列中； 对所述一级队列中的三类业务进行弹性分组环上环调度。
所述每个用户侧端口允许进入弹性分组环环路的 A、 B、 C三类业务的最小 带宽由用户配置。
所述业务等级调度是按照 A类业务优先 B类业务， B类业务优先 C类业务 的优先级顺序进行调度； 所述端口加权调度是根据各用户侧端口的各类业务的 权值进行调度。
所述权值是才艮据用户侧端口的 A、 B、 C三类业务的最小带宽占弹性分组环 设备的所有用户侧端口同类业务带宽的比值得到。
所述不同类业务等级的权值可以不相同。
所述对二级队列进行两级调度的步骤具体包括：
步骤 1) 依次查询用户侧端口的 A业务等级的二级队列是否为空， 如果为 空， 则执行步骤 2); 如果非空， 则转至步骤 3);
步骤 2) 判断该用户侧端口是否是最后一个用户侧端口， 如果是， 则转至 步 5); 如果不是， 则返回步骤 1), 查询下一个用户侧端口的 A业务等级的二 级队列；
步骤 3 ) 根据该用户侧端口 A类业务的权值， 将二级队列中的 A类业务调 度到 A类业务等级的一级队列内；
步骤 4 ) 判断 A类业务等级的一级队列是否满，如杲未满，则转至步骤 2 ); 如果已满， 则转至步骤 5);
步骤 5) 依次查询用户侧端口的 B业务等级的二级队列是否为空， 如果为 空， 则执行步骤 6); 如果非空， 则转至步骤 7);
步骤 6) 判断该用户侧端口是否是最后一个用户侧端口， 如杲是， 则转至 步骤 9); 如果不是， 则返回步骤 5)， 查询下一个用户侧端口的 B业务等级的二 级队列；
步骤 7 ) 根据该用户侧端口 B类业务的权值， 将二级队列中的 B类业务调 度到 B类业务等级的一级队列内；
步骤 8 ) 判断 B类业务等级的一级队列是否满，如果未满，则转至步骤 6 ); 如果已满， 则转至步骤 9 );
步骤 9 ) 查询用户侧端口的 C业务等级的二级队列是否为空， 如果为空， 则执行步骤 10 ); 如果非空， 则转至步骤 11 );
步骤 10 )判断该用户侧端口是否是最后一个用户侧端口， 如果是， 则转至 步骤 1 ); 如果不是， 则返回步骤 9 )， 查询下一个用户侧端口的 C业务等级的二 级队列；
步骤 11 ) 居该用户侧端口 C类业务的权值， 将二级队列中的 C类业务调 度到 C类业务等级的一级队列内；
步骤 12 ) 判断 C类业务等级的一级队列是否满， 如果未满， 则转至步骤 10 ); 如果已满， 则转至步骤 1 )。
所述对一级队列中三类业务进行上环调度的步骤进一步包括： 对三类业务 分别进行整形， 然后调度各类业务数据帧， 输出到相应的弹性分组环环路。
所述整形和调度是按照 A类业务优先 B类业务， B类业务优先 C类业务的 顺序进行的。
所述整形是根据各类业务的信用值对一级队列中的相应类业务进行整形。 所述各类业务的信用值是由弹性分组环设备允许进入弹性分组环环路的 A、 B、 C类业务的总带宽决定。
所述单个弹性分组环设备允许进入弹性分组环环路的 A、 B、 C类业务的总 带宽由用户配置。
所述上环调度的步骤具体包括：
步骤 1 ) 居 A类业务的信用值对一级队列中的 A类业务进行整形， 消耗 相应的信用值；
步骤 2 ) 调度整形后的 A类业务数据帧， 输出到弹性分组环环路； 步骤 3 ) 判断 A类业务的信用值是否消耗尽， 如果没有消耗尽， 则返回步 骤 1 ); 如果已经消耗尽， 则暂停对 A类业务的整形和调度， 执行步骤 4 );
步骤 4 ) 根据 B类业务的信用值对一级队列中的 B类业务进行整形， 消耗 相应的信用值；
步骤 5 ) 调度整形后的 B类业务数据帧， 输出到弹性分组环环路； 步骤 6 ) 判断 B类业务的信用值是否消耗尽， 如果没有消耗尽， 则返回步 骤 4 ); 如果已经消耗尽， 则暂停对 B类业务的整形和调度， 执行步骤 7 );
步骤 7 ) 根据 C类业务的信用值对一级队列中的 C类业务进行整形， 消耗 相应的信用值；
步骤 8 ) 调度整形后的 C类业务数据帧， 输出到弹性分组环环路； 步骤 9 ) 判断 C类业务的信用值是否消耗尽， 如杲没有消耗尽， 则返回步 骤 7 ); 如果已经消耗尽， 则暂停对 C类业务的整形和调度， 返回步骤 1 )。
本发明由于在每个用户侧端口配置了 A、 B、 C三个业务等级的二级队列， 并且提供了相应的对二级队列进行先业务等级调度后端口加权调度的两级调度 机制， 因此能够满足各个端口进入的 A类业务最先调度、 B类业务次之、 C类业 务最后调度的优先级次序， 从而保证了 A类业务和 B类业务对时延和时延抖动 的要求， 同时也使各个端口在同一等级的业务情况下不会出现饥饿现象。 本发 明方法有效地解决了存在多个用户端口的业务需要进入弹性分组环环路， 并且 每个用户端口同时存在 、 B、 C三类业务情况下的调度策略。
附图说明
图 1为弹性分组环设备的示意图； 图 2为现有技术中关于业务调度的框架示意图；
图 3本发明多级调度方法的框架示意图；
图 4是本发明中接收业务数据帧的流程图；
图 5是本发明中对各业务的二级队列进行二次调度的流程图；
图 6是本发明中进行上环调度的流程图。
具体实施方式
下面 居附图和实施例对本发明的技术方案做进一步的详细说明。
图 1和图 2是对现有技术的介绍， 已在前面背景技术部分详述过， 此处不 再赘述。
如图 3所示， 在弹性分组环设备的每个用户侧端口配置三个队列， 作为 A、 B、 C三类业务等级的二级队列。 用户侧端口接收到的数据帧， 首先进入相应业 务等级的二级队列中， 对二级队列中的各类业务数据帧进行先业务等级调度后 端口加权调度的两级调度， 经过两级调度的数据帧再进入到 A、 B、 C业务等级 的一级队列中， 再经过整形器的整形， 送入待发送队列中， 最后进入弹性分组 环环路。
单个弹性分组环设备允许进入弹性分组环环路的 A类业务的总带宽、 B类 业务的总带宽和 C类业务的总带宽， 以及其中每个用户侧端口允许进入弹性分 组环环路的 A类业务最小带宽、 B类业务最小带宽和 C类业务最小带宽， 均可 以由用户配置， 其中每个用户侧端口允许的 A类业务最小带宽的和与该弹性分 组环设备的 A类业务的总带宽是相等的， 对于 B类业务和 C类业务也有上述对 应关系。
用户侧端口接收数据帧的流程如图 4所示， 各用户侧端口接收数据帧， 然 后对接收到的数据帧根据其分类标识进行 A、 B、 C业务等级分类， 同时将数据 帧分别放入该端口所属的对应业务等级的二级队列中。
两级调度的流程如图 5所示。 ϋ殳单个弹性分组环站点有 Ν个用户侧端口 , 首先进行第一级的业务等级调度， 即按照先调度 Α类业务， 再调度 B类业务， 最后调度 C类业务的顺序依次进行。首先依次轮询用户侧端口 1 - N的 A类业务 二级队列， 如果被轮询到的当前用户侧端口的 A类业务二级队列为空， 则继续 轮询下一个用户侧端口的 A类业务二级队列， 否则 据该用户侧端口 A类业务 的权值将该用户侧端口的 A类业务二级队列里的业务数据帧调度到 A类业务的 一级队列中去。 A类业务的权值是 居用户配置的该用户侧端口的 A类业务带 宽占整个弹性分组环设备的 A类业务带宽的比例得到。 如果所有 A类业务等级 的二级队列都被调度空， 或者 A类业务等级的一级队列满， 则进行 B类业务等 级二级队列的调度。 B类业务等级二级队列的调度方法与上述 A类业务等级二 级队列的调度方法相同。 如杲所有端口的 B类业务等级的二级队列都被调度空 或者 B类业务等级的一级队列满， 则进行 C类业务等级二级队列的调度， C类 业务等级二级队列的调度方法与上述 A类业务等级二级队列的调度方法相同 , 上述步骤周而复始地循环。 需要说明的是， A、 B、 C 三类业务等级的端口权值 可以是不^！同的值。
图 6给出了上环调度的流程图， 对一级队列内的业务进行正常的弹性分组 环上环调度， 包括按照先 A类业务等级、 再 B类业务等级、 最后 C类业务等级 的优先级次序对业务数据的整形操作和调度过程。 根据标准规定， A 类业务又 分成了 AO类和 A1类， 其中 AO类业务的级别较高， 因此调度 A类业务时， 又按 照先 AO类业务， 再 A1类业务的顺序进行调度。 当调度 AO类业务时， 首先根据 AO的信用值对一级队列的 A类业务进行整形， 并消耗相应的 AO的信用值， 整 形后的数据帧发送到待发送队列，进入弹性分组环。当 AO的信用值被消耗尽后， 进行 Al类业务的调度， 才艮据 A1的信用值对一级队列的 A类业务进行整形， 并 消耗相应的 A1的信用值， 整形后的数据帧再通过 D整形器， 并消耗掉相应的 D 的信用值， 整形后的数据发送到待发送队列， 进入弹性分组环。 当 A1的信用值 被消耗尽后， 就暂停调度 A类业务， 转而调度 B类业务， 然后依次调度 C类业 务， 调度 B类、 C类业务的流程与 A1类业务类似。 不同类业务的信用值是由单 个弹性分组环设备允许进入弹性分组环环路的 A、 B、 C类业务的总带宽决定的。
虽然本发明的业务类别分类方法已被说明和描述， 但很明显本发明是不受 限制的。 在不偏离由附属权利要求书所确定的本发明的精神和范围的条件下， 本领域的技术人员将会考虑到许多修正、 更换、 变化、 替代和等效的内容。

Claims

权利要求书
1、 一种支持多端口多业务的多级调度方法， 其特征在于， 包括以下步骤： 在单个弹性分组环设备的每个用户侧端口配置 A、 B、 C业务等级的二级队 列 ,所述用户侧端口接收的数据帧根据业务分类标识存放在相应的二级队列中； 对所述用户側端口的二级队列进行先业务等级调度后端口加权调度的两级 调度；
经过两级调度的数据帧分别进入 A、 B、 C业务等级的一级队列中； 对所述一级队列中的三类业务进行弹性分组环上环调度。
2、才艮据权利要求 1所述的方法， 其特征在于： 所述每个用户侧端口允许进 入弹性分组环环路的 A、 B、 C三类业务的最小带宽可由用户配置。
3、 根据权利要求 1所述的方法， 其特征在于： 所述业务等级调度是按照 A 类业务优先于 B类业务， B类业务优先于 C类业务的优先级顺序进行调度； 所 述端口加权调度是根据各用户侧端口的各类业务等级的权值进行调度。
4、 根据权利要求 3所述的方法， 其特征在于： 所述权值是根据用户侧端口 的 λ、 B、 C三类业务的最小带宽占弹性分组环设备的所有用户侧端口同类业务 带宽的比值得到。
5、 据权利要求 4所述的方法， 其特征在于： 所述不同类业务等级的权值 可以不相同。
6、根据权利要求 1所述的方法， 其特征在于： 所述对二级队列进行两级调 度的步骤具体包括：
步骤 1 ) 依次查询用户侧端口的 A业务等级的二级队列是否为空， 如果为 空， 则执行步骤 2); 如果非空， 则转至步骤 3);
步骤 2) 判断该用户侧端口是否是最后一个用户侧端口， 如果是， 则转至 步骤 5); 如果不是， 则返回步骤 1)， 查询下一个用户侧端口的 A业务等级的二 级队列；
步骤 3 ) 根据该用户侧端口 A类业务的权值， 将二级队列中的 A类业务调 度到 A类业务等级的一级队列内；
步骤 4 ) 判断 A类业务等级的一级队列是否满，如果未满，则转至步驟 2 ); 如杲已满， 则转至步骤 5);
步骤 5) 依次查询用户侧端口的 B业务等级的二级队列是否为空， 如杲为 空， 则执行步骤 6); 如果非空， 则转至步骤 7);
步骤 6) 判断该用户侧端口是否是最后一个用户侧端口， 如果是， 则转至 步骤 9); 如杲不是， 则返回步骤 5)， 查询下一个用户侧端口的 B业务等级的二 级队列；
步骤 7 ) 才艮据该用户侧端口 B类业务的权值， 将二级队列中的 B类业务调 度到 B类业务等级的一级队列内；
步骤 8 ) 判断 B类业务等级的一级队列是否满，如果未满，则转至步骤 6 ); 如果已满， 则转至步骤 9);
步骤 9) 查询用户侧端口的 C业务等级的二级队列是否为空， 如果为空， 则执行步骤 10); 如果非空， 则转至步骤 11);
步骤 10)判断该用户侧端口是否是最后一个用户侧端口， 如果是， 则转至 步骤 1); 如果不是， 则返回步骤 9), 查询下一个用户侧端口的 C业务等级的二 级队列；
步骤 11 )根据该用户侧端口 C类业务的权值， 将二级队列中的 C类业务调 度到 C类业务等级的一级队列内；
步骤 12) 判断 C类业务等级的一级队列是否满， 如果未满， 则转至步骤 10); 如果已满， 则转至步骤 1)。
7、才 据权利要求 1所述的方法， 其特征在于： 所述对一级队列中三类业务 进行上环调度的步骤进一步包括： 对三类业务分别进行整形， 然后调度各类业 务数据帧， 输出到相应的弹性分组环环路。
8、 才艮据权利要求 7 所述的方法， 其特征在于： 所述整形和调度是按照 A 类业务优先 B类业务， B类业务优先 C类业务的顺序进行的。
9、才 M居权利要求 8所述的方法，其特征在于：所述上环调度的步骤具体为： 步骤 1) 根据 A类业务的信用值对一级队列中的 A类业务进行整形， 消耗 相应的信用值；
步骤 2) 调度整形后的 A类业务数据帧， 输出到弹性分组环环路； 步骤 3) 判断 A类业务的信用值是否消耗尽， 如果没有消耗尽， 则返回步 骤 1); 如果已经消耗尽， 则暂停对 A类业务的整形和调度， 执行步骤 4);
步骤 4) 根据 B类业务的信用值对一级队列中的 B类业务进行整形， 消耗 相应的信用值；
步骤 5) 调度整形后的 B类业务数据帧， 输出到弹性分组环环路； 步骤 6) 判断 B类业务的信用值是否消耗尽， 如果没有消耗尽， 则返回步 骤 4); 如果已经消耗尽， 则暂停对 B类业务的整形和调度， 执行步骤 7);
步骤 7) 根据 C类业务的信用值对一级队列中的 C类业务进行整形， 消耗 相应的信用值；
步骤 8) 调度整形后的 C类业务数据帧， 输出到弹性分组环环路； 步骤 9 ) 判断 C类业务的信用值是否消耗尽， 如果没有消耗尽， 则返回步 骤 7 ); 如果已经消耗尽， 则暂停对 C类业务的整形和调度， 返回步骤 1 )。
10、 ^^据权利要求 9所述的方法， 其特征在于： 所述各类业务的信用值是 由弹性分组环设备允许进入弹性分组环环路的 A、 B、 C类业务的总带宽决定。
11、 居权利要求 10所述的方法， 其特征在于： 所述单个弹性分组环设备 允许进入弹性分组环环路的 A、 B、 C类业务的总带宽由用户配置。
PCT/CN2004/000466 2003-08-15 2004-05-10 A method supporting the multilevel schedule of multi-port and multi-traffic Ceased WO2005018154A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN03140203.8 2003-08-15
CNB031402038A CN1298139C (zh) 2003-08-15 2003-08-15 一种支持多端口多业务的多级调度的方法

Publications (1)

Publication Number Publication Date
WO2005018154A1 true WO2005018154A1 (en) 2005-02-24

Family

ID=34155247

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2004/000466 Ceased WO2005018154A1 (en) 2003-08-15 2004-05-10 A method supporting the multilevel schedule of multi-port and multi-traffic

Country Status (3)

Country Link
CN (1) CN1298139C (zh)
RU (1) RU2350027C2 (zh)
WO (1) WO2005018154A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
RU2384950C2 (ru) * 2005-08-25 2010-03-20 ЭлДжи ЭЛЕКТРОНИКС ИНК. Способ предоставления услуги по совместно используемому каналу прямой линии связи
US20150058485A1 (en) * 2012-04-12 2015-02-26 Zte Corporation Flow scheduling device and method

Families Citing this family (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN100384182C (zh) * 2005-02-24 2008-04-23 杭州华三通信技术有限公司 在分组环上传送报文的方法
CN100474845C (zh) * 2005-10-28 2009-04-01 上海交通大学 对集成服务模型进行接纳控制的方法
CN100493019C (zh) * 2005-10-29 2009-05-27 华为技术有限公司 一种在环状网络中实现环向配置的方法和装置
CN102084628B (zh) * 2008-04-24 2014-12-03 马维尔国际有限公司 通信量管理器和用于通信量管理器的方法
CN101499973B (zh) * 2009-03-11 2012-05-23 中兴通讯股份有限公司 一种交换机实现全线速调度业务流的方法
CN101594307B (zh) * 2009-06-30 2012-12-05 中兴通讯股份有限公司 基于多级队列的调度方法和系统
CN101784082A (zh) 2009-12-22 2010-07-21 中兴通讯股份有限公司 无线局域网内增强服务质量的方法及装置
CN102025639A (zh) * 2010-12-23 2011-04-20 北京星网锐捷网络技术有限公司 队列调度方法及系统
CN102594662A (zh) * 2012-01-20 2012-07-18 中兴通讯股份有限公司 服务质量的实现方法及装置
CN102801626B (zh) * 2012-08-17 2015-02-11 北京空间飞行器总体设计部 一种基于aos标准的星载数据管理方法及装置
CN104410584A (zh) * 2014-12-16 2015-03-11 上海斐讯数据通信技术有限公司 一种QoS的队列等级调度方法
CN115242647A (zh) * 2021-04-06 2022-10-25 中国电信股份有限公司 用于动态流量整形的方法和装置

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1357188A (zh) * 1998-12-22 2002-07-03 鲍尔·埃克斯有限公司 数据交换方法和装置
WO2002080475A2 (en) * 2001-03-30 2002-10-10 Nokia Intelligent Edge Routers Inc. Method and apparatus for improved queuing
CA2255418C (en) * 1998-12-07 2003-01-21 Pmc-Sierra Ltd. Ring interface and ring network bus flow control system

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5751709A (en) * 1995-12-28 1998-05-12 Lucent Technologies Inc. Adaptive time slot scheduling apparatus and method for end-points in an ATM network
US6031832A (en) * 1996-11-27 2000-02-29 Telefonaktiebolaget L M Ericsson (Publ) Method and apparatus for improving performance of a packet communications system
GB2355623B (en) * 1999-10-19 2003-07-16 Ericsson Telefon Ab L M Packet transmission in a UMTS network
US6882623B1 (en) * 2000-02-08 2005-04-19 Native Networks Technologies Ltd. Multi-level scheduling method for multiplexing packets in a communications network
CN1165184C (zh) * 2001-05-08 2004-09-01 华为技术有限公司 一种适合实现路由器服务质量的综合业务中的调度方法
CN1184835C (zh) * 2002-01-21 2005-01-12 华为技术有限公司 无线分组数据传输系统的下行数据调度方法

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CA2255418C (en) * 1998-12-07 2003-01-21 Pmc-Sierra Ltd. Ring interface and ring network bus flow control system
CN1357188A (zh) * 1998-12-22 2002-07-03 鲍尔·埃克斯有限公司 数据交换方法和装置
WO2002080475A2 (en) * 2001-03-30 2002-10-10 Nokia Intelligent Edge Routers Inc. Method and apparatus for improved queuing

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
RU2384950C2 (ru) * 2005-08-25 2010-03-20 ЭлДжи ЭЛЕКТРОНИКС ИНК. Способ предоставления услуги по совместно используемому каналу прямой линии связи
US8861417B2 (en) 2005-08-25 2014-10-14 Lg Electronics Inc. Method of providing a service on a downlink shared channel
US9554357B2 (en) 2005-08-25 2017-01-24 Lg Electronics Inc. Method of providing a service on a downlink shared channel
US20150058485A1 (en) * 2012-04-12 2015-02-26 Zte Corporation Flow scheduling device and method
US10263906B2 (en) * 2012-04-12 2019-04-16 Zte Corporation Flow scheduling device and method

Also Published As

Publication number Publication date
RU2006107985A (ru) 2006-08-10
RU2350027C2 (ru) 2009-03-20
CN1581790A (zh) 2005-02-16
CN1298139C (zh) 2007-01-31

Similar Documents

Publication Publication Date Title
US9871739B2 (en) Service interface for QOS-driven HPNA networks
US7656798B1 (en) Service interface for QoS-driven HPNA networks
US6934250B1 (en) Method and apparatus for an output packet organizer
WO2005018154A1 (en) A method supporting the multilevel schedule of multi-port and multi-traffic
CN107733689A (zh) 基于优先级的动态加权轮询调度策略方法
WO2008095397A1 (en) Traffic scheduling method and apparatus thereof
CN101471854A (zh) 一种转发报文的方法及装置
CN101286949A (zh) 基于IEEE802.16d标准的无线Mesh网MAC层资源调度策略
CN101834787A (zh) 调度数据的方法和系统
CN101692648A (zh) 一种队列调度方法及系统
JP2006506845A (ja) ルータにおけるパケットに対し論理リンクを選択する方法
CN103905336A (zh) 一种基于网络编码技术的SIEPON中QoS感知的调度方法
CN100420249C (zh) 一种保证无线局域网中的服务质量方法
CN100466593C (zh) 一种支持多业务的综合队列调度的实现方法
JP3623420B2 (ja) トラヒック制御方法
CN101616096B (zh) 队列调度方法及装置
CN106789716A (zh) Tdma自组网的mac层队列调度方法
CN114257554B (zh) 一种提高tsn网络be流带宽利用率的调度方法
CN105099947B (zh) 空间网络接入方法及装置
Chen et al. Packet scheduling algorithm based on priority adjustment in wireless sensor networks
US6904056B2 (en) Method and apparatus for improved scheduling technique
CN108347378A (zh) 一种用于大电网的控制专用网络及动态路由方法
CN117240795A (zh) 一种支持多模态处理逻辑的调度方法及系统
Sheu et al. A generalized channel preemption model for multiclass traffic in mobile wireless networks
CN101188547A (zh) 结合虚拟监控组和car速率限制提高转发效率的路由器

Legal Events

Date Code Title Description
AK Designated states

Kind code of ref document: A1

Designated state(s): AE AG AL AM AT AU AZ BA BB BG BR BW BY BZ CA CH CN CO CR CU CZ DE DK DM DZ EC EE EG ES FI GB GD GE GH GM HR HU ID IL IN IS JP KE KG KP KR KZ LC LK LR LS LT LU LV MA MD MG MK MN MW MX MZ NA NI NO NZ OM PG PH PL PT RO RU SC SD SE SG SK SL SY TJ TM TN TR TT TZ UA UG US UZ VC VN YU ZA ZM ZW

AL Designated countries for regional patents

Kind code of ref document: A1

Designated state(s): GM KE LS MW MZ NA SD SL SZ TZ UG ZM ZW AM AZ BY KG KZ MD RU TJ TM AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HU IE IT LU MC NL PL PT RO SE SI SK TR BF BJ CF CG CI CM GA GN GQ GW ML MR NE SN TD TG

121 Ep: the epo has been informed by wipo that ep was designated in this application
WWE Wipo information: entry into national phase

Ref document number: 288/MUMNP/2006

Country of ref document: IN

WWE Wipo information: entry into national phase

Ref document number: 2006107985

Country of ref document: RU

122 Ep: pct application non-entry in european phase