[go: up one dir, main page]

WO2014180447A1 - 一种带宽地图更新的方法及装置 - Google Patents

一种带宽地图更新的方法及装置 Download PDF

Info

Publication number
WO2014180447A1
WO2014180447A1 PCT/CN2014/079966 CN2014079966W WO2014180447A1 WO 2014180447 A1 WO2014180447 A1 WO 2014180447A1 CN 2014079966 W CN2014079966 W CN 2014079966W WO 2014180447 A1 WO2014180447 A1 WO 2014180447A1
Authority
WO
WIPO (PCT)
Prior art keywords
status table
resource status
wavelength
node
bandwidth
Prior art date
Application number
PCT/CN2014/079966
Other languages
English (en)
French (fr)
Inventor
王磊
严芬
施社平
任之良
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
Priority to EP14794041.5A priority Critical patent/EP3038279B1/en
Priority to JP2016535314A priority patent/JP6373996B2/ja
Priority to US14/912,930 priority patent/US9698930B2/en
Publication of WO2014180447A1 publication Critical patent/WO2014180447A1/zh

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04JMULTIPLEX COMMUNICATION
    • H04J14/00Optical multiplex systems
    • H04J14/02Wavelength-division multiplex systems
    • H04J14/0227Operation, administration, maintenance or provisioning [OAMP] of WDM networks, e.g. media access, routing or wavelength allocation
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04JMULTIPLEX COMMUNICATION
    • H04J14/00Optical multiplex systems
    • H04J14/02Wavelength-division multiplex systems
    • H04J14/0278WDM optical network architectures
    • H04J14/0283WDM ring architectures
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04JMULTIPLEX COMMUNICATION
    • H04J14/00Optical multiplex systems
    • H04J14/02Wavelength-division multiplex systems
    • H04J14/0201Add-and-drop multiplexing
    • H04J14/0215Architecture aspects
    • H04J14/022For interconnection of WDM optical networks
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04JMULTIPLEX COMMUNICATION
    • H04J14/00Optical multiplex systems
    • H04J14/02Wavelength-division multiplex systems
    • H04J14/0227Operation, administration, maintenance or provisioning [OAMP] of WDM networks, e.g. media access, routing or wavelength allocation
    • H04J14/0254Optical medium access
    • H04J14/0256Optical medium access at the optical channel layer
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04JMULTIPLEX COMMUNICATION
    • H04J14/00Optical multiplex systems
    • H04J14/02Wavelength-division multiplex systems
    • H04J14/0227Operation, administration, maintenance or provisioning [OAMP] of WDM networks, e.g. media access, routing or wavelength allocation
    • H04J14/0254Optical medium access
    • H04J14/0267Optical signaling or routing
    • H04J14/0269Optical signaling or routing using tables for routing
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04JMULTIPLEX COMMUNICATION
    • H04J14/00Optical multiplex systems
    • H04J14/02Wavelength-division multiplex systems
    • H04J14/0227Operation, administration, maintenance or provisioning [OAMP] of WDM networks, e.g. media access, routing or wavelength allocation
    • H04J14/0254Optical medium access
    • H04J14/0272Transmission of OAMP information
    • H04J14/0273Transmission of OAMP information using optical overhead, e.g. overhead processing
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04QSELECTING
    • H04Q11/00Selecting arrangements for multiplex systems
    • H04Q11/0001Selecting arrangements for multiplex systems using optical switching
    • H04Q11/0062Network aspects
    • H04Q11/0066Provisions for optical burst or packet networks
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04QSELECTING
    • H04Q11/00Selecting arrangements for multiplex systems
    • H04Q11/0001Selecting arrangements for multiplex systems using optical switching
    • H04Q11/0062Network aspects
    • H04Q2011/0064Arbitration, scheduling or medium access control aspects
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04QSELECTING
    • H04Q11/00Selecting arrangements for multiplex systems
    • H04Q11/0001Selecting arrangements for multiplex systems using optical switching
    • H04Q11/0062Network aspects
    • H04Q2011/0086Network resource allocation, dimensioning or optimisation
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04QSELECTING
    • H04Q11/00Selecting arrangements for multiplex systems
    • H04Q11/0001Selecting arrangements for multiplex systems using optical switching
    • H04Q11/0062Network aspects
    • H04Q2011/009Topology aspects
    • H04Q2011/0092Ring

Definitions

  • the present invention relates to the field of optical network technologies, and more particularly to a wavelength-time slot allocation and bandwidth map updating method and apparatus for a sub-wavelength all-optical switching network, that is, an Optical Burst Transport Network (OBTN).
  • OBTN Optical Burst Transport Network
  • the flexible pipelines and statistical multiplexing characteristics of packet-switched networks are naturally adapted to data.
  • the current packet switching is basically based on electrical layer processing, which has high cost and high energy consumption. With the rapid growth of traffic, the processing bottleneck is becoming more and more prominent, and it is difficult to adapt to the needs of future high speed, flexible, low cost and low energy consumption of the network. .
  • Optical networks have the advantages of low cost, low power consumption, and high speed and large capacity.
  • traditional optical circuit switching networks such as WDM (Wavelength Division Multiplexing) and OTN (Optical Transport Network) can only provide large
  • WDM Widelength Division Multiplexing
  • OTN Optical Transport Network
  • the GPON Gigabit-Capable Passive Optical Network
  • the downlink signal transmitted by the OLT is distributed to each ONU (Optical Network Unit) through the optical splitter by means of optical layer broadcasting.
  • the downlink frame header carries a bandwidth map of the uplink frame to indicate the transmission time and length of each ONU uplink data.
  • each ONU sends data according to the bandwidth map indication, and passes through the optical coupler. Multiplexed to one wavelength channel and uploaded to the OLT. In this way, GPON has the characteristics of high speed, large capacity and low cost of the optical layer.
  • GPON generally adopts a star/tree network topology, and its working principle is suitable for carrying multi-point to single-point aggregation traffic (the north-south traffic dominates), so it is successfully applied and large-scale deployment in the access network.
  • OBTN Optical Burst Transport Network
  • FIG. 1 is a schematic diagram of a 4-node OBTN unidirectional ring network.
  • Each node is equipped with a pair of fast tunable burst transmitters and fast tunable burst receivers (which can be expanded into multiples).
  • the whole network has two wavelengths as data channels. , one wavelength is used as the control channel, and node A is the master node.
  • the technical characteristics of OBTN are briefly described as follows:
  • the most basic transmission unit in the data channel is OB. There is guard time between OBs as intervals. One or more OBs form a data frame. The corresponding OB frames and OB slot start positions of different wavelength channels need to be aligned.
  • the data channel uses a burst optical receiver/transmitter. The burst data is directly transmitted by the optical layer between the source and sink nodes, and no intermediate node is required for electrical layer forwarding.
  • the source side needs to aggregate and encapsulate the client side data packets to the OB to send.
  • the control channel is separated from the data channel.
  • OBTN uses independent wavelength channel to carry control information, including OAM (Operation Administration and Maintenance) information, bandwidth report for collecting bandwidth requests of each node, and bandwidth map indicating each node to send/receive data, and control
  • the frame is sent before the corresponding data frame.
  • the control channel can use a common optical receiver/transmitter as a transceiver device, and each domain performs electrical domain processing to receive and update corresponding control information.
  • the timing relationship between the control frame and the data frame is shown in Figure 2.
  • the OBTN node can quickly adjust (ns (nanosecond) level) transmitter/receiver transmit/receive wavelengths to map based on bandwidth The corresponding wavelength and OB slot are selected for burst data transmission/reception to implement OB-based all-optical switching.
  • OBTN adopts centralized control mode.
  • Each slave node periodically reports bandwidth requests to the master node through the control frame.
  • the master node performs wavelength and OB time slot allocation according to the current resource status and bandwidth allocation policy, and records the allocation result into the bandwidth map.
  • the control frame is distributed to each slave node to implement fast scheduling of optical layer resources according to traffic requirements.
  • the burst data packet is directly transmitted to the optical layer of the source-sink node, it is not subjected to electrical processing, and thus is constrained by wavelength consistency and slot consistency.
  • a burst data packet sent by node A to node D is specified by the bandwidth map on the third ⁇ time slot of the corresponding data frame of wavelength ⁇ 1 , because the optical layer transmits directly, and there is no wavelength conversion.
  • optical buffer wavelength converter is costly and has a serious impact on signal quality; optical buffer technology is not broken.
  • this burst packet When passing through node ⁇ and node C to node D, this burst packet must also occupy wavelength ⁇ 1 correspondingly.
  • the third ⁇ time slot of the data frame, and the wavelength and time slot positions cannot be changed.
  • the resource conflicts in Suizhong mainly include the following three types:
  • a source transmitter can transmit burst data at only one wavelength at any slot position. As shown in the figure, node ⁇ transmits a burst of A-->D in the third ⁇ slot of the corresponding data frame of wavelength ⁇ ;; at this time, if there is another service in the bandwidth map with the node A as the source node A transmitter resource collision occurs on the third one of the corresponding data frames of wavelength ⁇ 2 .
  • a destination receiver can receive burst data at only one wavelength at any slot position. As shown in the figure, node D receives a burst of ->0 in the third ⁇ time slot of the corresponding data frame of wavelength ⁇ ;; at this time, if there is another service in the bandwidth map with the D node as the destination node, A receiver resource conflict occurs when the wavelength ⁇ 2 corresponds to the third ⁇ time slot of the corresponding data frame.
  • Embodiments of the present invention provide a method and apparatus for updating a bandwidth map to eliminate resource conflicts in a bandwidth map calculation and update process.
  • the master node After receiving the bandwidth report carried by the control frame, the master node creates a new bandwidth map, creates a new resource status table, and sets all resource states in the newly created resource status table to be available.
  • the map is distributed hop by hop to the slave node.
  • the foregoing method further has the following features: the transmitting a channel allocation structure across the primary node according to the bandwidth map to be updated, and adding the newly created bandwidth map to the lower node allocation structure of the transmission channel across the primary node, including:
  • the newly created resource status table includes: a source/sink resource status table of a new allocation period, a link resource status table of a current allocation period, and a sink resource status table of a next allocation period. .
  • the above method has the following features:
  • the total number of gaps is an LXWXM three-dimensional matrix table, indicating an available state of
  • the method further has the following features: the updating the corresponding status table includes: the updated resource status table includes:
  • the foregoing method further has the following features: the allocating the wavelength and the optical burst time slot one by one for the current bandwidth request, including:
  • the source resource state table and the sink resource state table Querying the source resource state table and the sink resource state table according to the source node identifier and the sink node identifier of the current bandwidth request, where the source resource state table and the sink resource state table have the same optical burst time slot available. And querying the routing table according to the source node identifier and the sink node identifier of the current bandwidth request. If there is an unselected route in the routing table, selecting a route, and querying the link resource according to the selected route according to the wavelength order.
  • the status table if the link has the same optical burst slot resource available at the same wavelength and the source end, select a wavelength that satisfies the requirement and the optical burst slot position for allocation.
  • the above method has the following features:
  • the new bandwidth map carries wavelength and optical burst slot configuration information for each burst transmission channel, the new bandwidth map consisting of one or more sub-maps, each sub-map comprising a wavelength field and an optical burst slot allocation Field.
  • the above method has the following features:
  • the wavelength field is composed of a wavelength identifier and a number of optical bursts, and the optical burst slot allocation field is in a bit mapping manner, and each bit corresponds to an optical burst slot position in the data frame.
  • the above method further has the following features: the carrying the new bandwidth by the control frame
  • the map is hop-by-hop distributed to the slave nodes, including:
  • Object encoding is performed according to the new bandwidth map, and the encoded new bandwidth map is carried by the control frame to the slave node hop by hop.
  • an embodiment of the present invention further provides an apparatus for updating a bandwidth map, where: an initialization module, an allocation module, a generation module, a distribution module, and an update module, where: the initialization module is configured to: receive a control frame After the bandwidth is carried, the new bandwidth map is created, and the new resource status table is created, and all resource statuses in the newly created resource status table are made available;
  • the allocation module is configured to: add a newly created bandwidth map across the primary node transmission channel drop allocation structure according to the bandwidth distribution map of the bandwidth map to be updated, and trigger the update module;
  • the generating module is configured to: allocate wavelength and optical burst time slots one by one according to the bandwidth report reported by the control frame, add the newly created bandwidth map, generate a new bandwidth map, and trigger the update module;
  • the distribution module is configured to: traversely distribute the new bandwidth map to the slave node by using a control frame;
  • the update module is configured to: update the resource status table after being triggered.
  • the above device also has the following features:
  • the allocation module is configured to transmit a channel-on-road allocation structure across the primary node according to the bandwidth map to be updated in the following manner, and add a new bandwidth map across the primary node transmission channel to the downlink allocation structure: traverse the bandwidth map to be updated, and Transmitting the wavelength of the channel allocation structure and the position of the optical burst slot across the primary node, and configuring the wavelength of the downlink allocation structure and the optical burst slot position of the same transmission channel in the newly created bandwidth map.
  • the above device also has the following features:
  • the initialization module is configured to newly create a resource status table in the following manner: a new source/sink resource status table of a current allocation period, a link resource status table of a current allocation period, and a sink resource status table of a next allocation period,
  • the number of OBTN ring network nodes, I indicates the number of node data channel transceivers, M indicates the total number of optical burst slots of a data frame;
  • the above device also has the following features:
  • the update module is configured to update the corresponding resource status table in a manner of: updating a source/sink resource status table of the current allocation period and a link resource status of the current allocation period for non-cross-node service Updating the source resource status table of the current allocation period, the link resource status table of the current allocation period, and the sink resource status table of the next allocation period for the cross-primary node uplink service;
  • the road service updates the sink resource status table of the current allocation period and the link resource status table of the current allocation period.
  • the above device also has the following features:
  • the generating module is configured to allocate wavelength and optical burst slots one by one for the current bandwidth request in the following manner: querying the source resource status table and the sink resource status table according to the source node identifier and the sink node identifier of the current bandwidth request, If the source resource status table and the sink resource status table have the same optical burst time slot, the routing table is queried according to the source node identifier and the sink node identifier of the current bandwidth request, as in the routing table. If the route is not selected, a route is selected, and the link resource state table is queried according to the selected route according to the wavelength order. For example, if the link has the same optical burst time slot resource at the same wavelength and the source end, then the route is selected. A wavelength that satisfies the requirements and the position of the optical burst slot are allocated.
  • the above device also has the following features:
  • the generating module generates a new bandwidth map carrying wavelength and optical burst slot configuration information of each burst transmission channel, where the new bandwidth map is composed of one or more submaps, each submap includes a wavelength field and a light a burst slot allocation field, the wavelength field is composed of a wavelength identifier and a number of optical bursts, and the optical burst slot allocation field is in a bit mapping manner, and each bit corresponds to an optical burst slot position in the data frame. .
  • the above device also has the following features:
  • the distribution module is configured to perform object coding according to the new bandwidth map, and the new bandwidth map carried by the control frame is hop-by-hop distributed to the slave node.
  • the embodiment of the invention further provides a computer program, comprising program instructions, when the program instruction is executed by the master node, so that the master node can execute the above method.
  • Embodiments of the present invention also provide a carrier carrying the above computer program.
  • an embodiment of the present invention provides a method and an apparatus for updating a bandwidth map.
  • resource conflicts are eliminated in a bandwidth map calculation and update process, and the bandwidth dynamic allocation of any node pair can be reduced while reducing the bandwidth.
  • Packet rate improve the carrying capacity of network services.
  • FIG. 1 is a schematic diagram of a basic structure of an OBTN ring network
  • FIG. 2 is a schematic diagram of relative timing of control frames and data frames in an OBTN ring network
  • FIG. 3 is a schematic diagram of an example of a source/sink resource status table according to an embodiment of the present invention.
  • FIG. 4 is a schematic diagram of an example of a link resource status table according to an embodiment of the present invention.
  • FIG. 5 is a schematic diagram of an example of a routing table according to an embodiment of the present invention.
  • FIG. 6 is a flowchart of a bandwidth map updating method according to an embodiment of the present invention.
  • FIG. 7 is a flowchart of a wavelength and time slot allocation method according to an embodiment of the present invention.
  • FIG. 8 is a schematic diagram of encoding a bandwidth map object according to an embodiment of the present invention.
  • FIGS. 9a to 9d are schematic diagrams of bandwidth map update according to an embodiment of the present invention.
  • FIG. 10a is an exemplary diagram of wavelengths and OB slot allocations for A->C of the network scenario of FIG. 1;
  • FIG. 10b is an exemplary diagram of wavelengths and OB slot allocations for C-->B for the network scenario of FIG. 1;
  • FIG. 11 is a schematic diagram of an apparatus for updating a bandwidth map according to an embodiment of the present invention. Preferred embodiment of the invention
  • OBTN has the advantages of full photon wavelength switching, but due to optical buffer technology and wavelength conversion
  • the immature technology makes it impossible to learn from the "storage-and-forward" method of using traditional electrical layering devices, which greatly increases the possibility of resource conflicts and may cause a large amount of packet loss. Therefore, it is necessary to consider the resource conflict problem in the bandwidth map calculation phase, through reasonable wavelength slot resource allocation, and combined with the corresponding bandwidth map update method to avoid resource conflicts and improve the efficiency of the OBTN network; meanwhile, the OBTN network needs to be implemented.
  • Fast resource scheduling so the complexity of the wavelength time slot allocation algorithm must be reasonable to meet real-time requirements.
  • this problem has not been considered in current OBTN related literature and patents.
  • the requirements of the OBTN networking scenario are as follows:
  • the ring topology is supported, which may be single fiber unidirectional or dual fiber bidirectional or multi-fiber bidirectional.
  • the nodes in the network can be configured with one or more pairs of data channel transceivers, and the data channel transceiver can adopt three methods: a. Fast tunable burst transmission Fast Tunable Burst Mode Transmitter (FT-BMT) + Normal Tuner (Fast Tunable Burst Mode Receiver, FT-BMR); b. Normal Transmitter + FT-BMR; c. FT-BMT+FT-B.
  • FT-BMT Fast Tunable Burst Mode Transmitter
  • Figure 1 shows the basic structure of the OBTN unidirectional ring network.
  • the network scenario is as follows:
  • the network topology is a 4-node unidirectional ring network, node A is the master node, and other nodes are slave nodes;
  • Each node configures a pair of fast tunable burst transmitters/receivers in the data channel, and configures a pair of ordinary transmitters/receivers in the control channel;
  • the ring network data channel is configured with 2 wavelengths, and the control channel is configured with one wavelength;
  • the data frame contains 10 OB slots, and the loop data loop length is 4 times the data frame length;
  • the control frame is generated by the master node, updated by the slave node, transmitted along the ring for one week and terminated at the master node;
  • the DBA (Dynamic Bandwidth Allocation) algorithm executes once for one data frame.
  • the embodiment is mainly described with reference to the scenario of FIG. 1, but in addition to the scenario, the present invention is also applicable to any one-way/two-way OBTN ring network scenario, and the star and tree topology are relatively simple, and the invention is The content of the embodiment can be simplified by applying it.
  • bandwidth map update in Figure 1 is mainly divided into the following four stages: 1) Bandwidth request reporting:
  • the control frame carries the bandwidth report to the master node, and the master node receives and quantizes it to each node to the OB time slot request;
  • the master node performs a DBA algorithm according to the current resource state and bandwidth request, performs wavelength and time slot allocation, and generates a new bandwidth map;
  • the control frame carries the bandwidth map, and is sent to the slave node hop by hop in advance according to the corresponding data frame, and the slave node receives and generates the node map and the downlink map of the node, and at the same time, the control frame collects the bandwidth request at each slave node;
  • 9a to 9d are embodiments of bandwidth map update for the network scenario of FIG. 1, and the detailed process is as follows:
  • the master node When the master node receives the bandwidth report of the control frame and enters the bandwidth map calculation process, it first performs the bandwidth map update, as shown in Figure 6, the steps are as follows:
  • Step 101 After receiving the bandwidth report reported by the control frame, the master node creates a new bandwidth map, creates a corresponding resource status table, and sets all resource states in the newly created table to be available.
  • the master node creates a new unallocated bandwidth map called a new bandwidth map.
  • the master node receives the control frame bandwidth report, enters the bandwidth map calculation process, and creates a new bandwidth map and the above three tables in the first step of initialization.
  • the lower periodic camp resource status table represents the fourth after the current data frame corresponding to the bandwidth map.
  • the sink resource status table of the data frame (because the OB time slot allocated in the current bandwidth map across the primary node's uplink service will pass through the primary node, and is regarded as the OB time slot of the lower periodic bandwidth map allocated across the primary node's downlink service, occupying
  • the substation resource table corresponding to the next periodic data frame and the link resource status table are set to 1 to indicate that the resource is not occupied.
  • Step 102 Add a new bandwidth map across the primary node transmission channel downlink allocation structure according to the cross-master node transmission channel uplink allocation structure of the bandwidth map to be updated, and update the current resource allocation status table and link according to the resource occupation.
  • Resource status table The master node traverses the bandwidth map to be updated, that is, the old bandwidth map, and configures the wavelength and OB slot position of the uplink allocation structure across the primary node to be the wavelength of the same transmission channel in the new bandwidth map. And the slot position, simultaneously updating the sink resource status table of the current allocation period and the link resource status table of the current allocation period.
  • the OB time slot allocated by the uplink traffic of the old bandwidth map across the primary node will occupy the corresponding link and the downlink resource within the effective period of the new bandwidth map after passing through the primary node, that is, it is regarded as a new bandwidth map.
  • the OB time slot allocated by the downlink service is transmitted across the primary node. Therefore, it is necessary to traverse the old bandwidth map.
  • the wavelength and OB slot position of the uplink allocation structure across the primary node transmission channel are configured to be the same in the new bandwidth map.
  • the uplink allocation structure of the transmission channel across the primary node includes: wavelength 1 D-->C uplink of the 5th OB slot, wavelength 2 2nd OB slot C-->B on the road, 3rd OB slots D-->B on the road, the fourth OB slot D-->C on the way, the sixth OB slot D-->C on the way.
  • the sink resource status table and the link resource status table of the current allocation period are updated according to the resource occupation status of the transmission channel drop allocation structure across the primary node.
  • the D-->C downlink of the 5th OB slot of wavelength 1 needs to occupy the C-node downlink resource of the 5th OB slot, and the link AB and BC are at the 5th OB slot of the wavelength 1
  • the gap resource, so its 0 flag is not available, and then this resource is not selected during wavelength and time slot allocation.
  • the final resource status table update results are shown in Figure 9b.
  • the bandwidth map to be updated represents the bandwidth map corresponding to the fourth data frame before the current data frame, because the OB slot allocated by the uplink service of the bandwidth map of the to-be-updated bandwidth map passes through the master node and is regarded as The OB slot allocated by the downlink service of the current bandwidth map of the current bandwidth map occupies the downlink channel resource corresponding to the current data frame. Because there is mutual influence between the front and rear periodic bandwidth maps, in this embodiment, the primary node needs to save at least four bandwidth maps.
  • Step 103 According to the bandwidth report reported by the control frame, allocate wavelength and OB time slots one by one for the current bandwidth request, add a new bandwidth map, and update the corresponding status table.
  • the source/sink resource status table and the link resource status table of the current allocation period need to be updated.
  • the source resource status table, the link resource status table, and the link resource status table of the current allocation period need to be updated.
  • the bandwidth request is quantized and the wavelength and OB time slot are allocated, and the corresponding resource status table is updated.
  • A-->C requests to allocate 1 OB time slot.
  • the second OB time slot of wavelength 1 is allocated to it.
  • the resource occupation is as follows: Node A 2nd OB
  • C ⁇ >B requests to allocate 1 OB slot, and after the state table query and the wavelength slot allocation algorithm, it allocates the 3rd OB slot of wavelength 2.
  • the resource occupancy is as follows: The uplink resources of the OB slots, the downlink resources of the third OB slot of the Node B in the next cycle, and the slot resources of the links CD and DA at the 3rd OB slot of the wavelength 2. Set the corresponding bit to 0 in the resource table.
  • the route of C-->B passes through the master node, and therefore only occupies the uplink resource of the current allocation period of the C node and the slot resources of the link CD and DA, and the downlink resource of the Node B and the link time slot resource of the link AB Occupied in the bandwidth map of the next allocation cycle.
  • the final bandwidth map and resource status table as shown in Figure 9c.
  • Step 104 The master node carries the new bandwidth map to the slave node hop by hop through the control frame.
  • the new bandwidth map is constructed, saved in the primary node, and the bandwidth map object is generated according to the bandwidth map coding manner, and the control frame carries the hop-by-hop distribution to the slave node.
  • the bandwidth map needs to be carried in the control frame to be passed to the slave node to guide which services the slave node is going on and off at which wavelengths and time slots. Therefore, the bandwidth map needs to have a consistent and standardized coding method to facilitate the master node coding, control frame carrying and slave node identification.
  • the bandwidth map object is encoded and carried to the slave node in the control frame, and the control frame and the data frame timing are as shown in FIG. 2; the slave node generates the transmit wavelength slot table and the receive wavelength slot according to the bandwidth map.
  • the table as shown in Figure 1, is validated by a certain delay (the node delay is determined by measurement and carried in the control frame, and its implementation mechanism is outside the scope of this patent) to control the data transmission and reception of the node.
  • the D-->C downlink occupies 3 OB slots in the bandwidth map, which is the 5th slot of the wavelength 1, respectively, which is reflected in the OB slot allocation code of the corresponding BTC (burst transmission channel) overhead.
  • 0x0010 the 5th bit is set
  • the 4th and 6th time slots of wavelength 2, in the OB time slot allocation code are represented as 0x0028 (the 4th and 6th bit are set).
  • D-->C is in the bandwidth map, it belongs to the distribution structure of the transmission channel across the primary node. It is only received within the effective range of the bandwidth map. Therefore, the location of the uplink is 0, and the location of the downlink is 1; The path allocation structure does not carry a bandwidth report; the source node of this BTC is indicated in the associated BMS (burst multiplex section) overhead.
  • A-->C belongs to the non-cross-master transmission channel in the bandwidth map, and is sent and received within the effective range of the bandwidth map. Therefore, both the upper and lower path bits are set to 1; the BTC overhead of the primary node is the source node. There is no need to carry a bandwidth report, which is directly available from local reads.
  • the C-->B on the bandwidth map belongs to the transmission structure of the transmission channel across the primary node, and only transmits within the effective range of the bandwidth map, so only the uplink location is 1; it may carry the bandwidth report, but the master node generates
  • the bandwidth map does not include the bandwidth report, but is carried by the control frame when it arrives at the slave node.
  • the slave node adds and reports the corresponding BTC overhead of the control frame according to requirements, as shown in Figure 9d.
  • source/sink resource state table The following state table is needed in the embodiment of the present invention: source/sink resource state table, link resource state table, and routing table:
  • the source/sink resource status table is an XI three-dimensional matrix, where N represents the number of OBTN ring network nodes, I represents the number of node data channel transceivers, and M represents the total number of OB slots of a data frame. Its Indicates the available state of the i-th transmitter/receiver of node n at the mth OB slot position, 1 is resource available, and 0 is resource occupancy.
  • the link resource status table is a L W X M three-dimensional matrix, where L represents the number of links, W represents the number of data channel wavelengths, and M represents the total number of OB slots of a data frame. It indicates the available state of the wavelength w in the link 1 at the mth OB slot position, 1 is resource available, and 0 is resource occupancy.
  • the routing table is a N X N X PXL four-dimensional matrix, where N represents the number of OBTN ring network nodes, P represents the number of routes of the node pair (s. d), and L represents the number of links. For a unidirectional ring, the value is 1; for a two-way ring, the value is 2. It indicates whether the route p of node s to node d passes through link 1, 1 is elapsed, and vice versa.
  • 3, 4, and 5 are examples of the source/sink resource status table, link resource status table, and routing table for the scenario of FIG. 1 in this embodiment, respectively.
  • the innovations and advantages of the bandwidth map update method in this embodiment are summarized as follows: (1)
  • the resource status table can effectively identify the resource occupation of the transmission channel across the primary node and the resource occupation of the currently allocated transmission channel.
  • the resource and the time slot are allocated without resource conflicts, so the generated bandwidth map does not have resource conflicts.
  • the bandwidth map update method considers the resource conflict problem in the wavelength and time slot allocation process, and can effectively improve the network.
  • the bearer capability and the high resource utilization rate; the existing bandwidth map update method detects the conflict after the bandwidth map is generated, and cancels the allocated transport channel related to the conflict resource in the bandwidth map to avoid the conflict, resulting in unstable bandwidth allocation and resource utilization.
  • the rate is low and network performance is difficult to guarantee.
  • the wavelength table and the OB time slot allocation method based on the state table are as shown in FIG. Step 201: Select a bandwidth request from the bandwidth request set.
  • Step 202 Query the source resource status table and the sink resource status table in sequence according to the source and sink node ID of the selected request.
  • the current allocation period source and sink resource status table For the non-cross-node service, query the current allocation period source and sink resource status table. For the cross-primary node service, query the current allocation period source resource status table and the next allocation period sink resource status table. If the source resource status table and the sink resource status table have the same OB time slot available, go to step 203; if the source and sink resource status table has no consistent time slot available after traversing the status table, the allocation fails, and the process goes to step 205.
  • Step 203 Query the routing table according to the source and sink node ID of the selected request. If there is no route selected, select the route, mark it as selected, go to step 204; if all routes have been traversed, and the allocation is not completed, Go to step 205.
  • Step 204 Query the link resource status table according to the wavelength according to the route. If the link has the same OB slot resource available at the same wavelength and the source end, select a wavelength that meets the requirement and the OB slot position for allocation. It is added to the bandwidth map, and the allocation is completed; if there is no consistent OB slot position for traversing each wavelength, then go to step 203.
  • Step 205 Remove the request from the bandwidth request set, and the wavelength and OB time slot allocation fails.
  • FIG. 10a is an exemplary diagram of wavelengths and OB slot allocations for A->C of the network scenario of FIG. 1
  • FIG. 10b is an exemplary diagram of wavelengths and OB slot allocations for C-->B for the network scenario of FIG.
  • one wavelength and OB time slot are selected from the available sets for allocation:
  • the seventh OB time slot of wavelength 1 is allocated.
  • one wavelength and OB time slot are selected from the available sets for allocation:
  • the fourth OB time slot of wavelength 2 is allocated.
  • it is added to the bandwidth map, and the current resource allocation status table of the current allocation period of the C node, the resource status table of the next allocation period of the B node, and the resource status table of the link CD and DA at the wavelength 2 are updated.
  • the innovations and advantages of the wavelength and time slot allocation method in this embodiment are as follows: (1) The resource state table can take a bit mapping manner, which can effectively reduce the storage space requirement; (2) the wavelength and time slot allocation based on the state table. The method only needs simple table lookup and OR operation, has low complexity, meets the requirements of real-time calculation and fast resource scheduling, and is suitable for hardware implementation; (3) The wavelength and time slot allocation method in this embodiment is suitable for all three OBTNs. Node structure; (4) There are currently no related patents and documents to solve this problem.
  • control frame bandwidth map object coding structure in this embodiment is as shown in FIG. 8. This application only pays attention to the coding structure related to the bandwidth map, and does not expand in detail for other object structures of the control frame.
  • the burst multiplex section (BMS) overhead is used to specify a node (source node) and carries the OAM overhead associated with this node.
  • the BMS overhead mainly includes three parts: a source node ID, a node OAM overhead, and a length indication.
  • the length indication is composed of a bandwidth map presence identifier, a burst transmission channel (BTC) number, and a CRC check.
  • the number of BTCs indicates the total number of BTCs with the node ID in the BMS as the source node, and the bandwidth map identifier indicates whether these BTCs carry the bandwidth map.
  • the control frame includes n burst multiplex section (BMS) overheads.
  • the burst transfer channel (BTC) overhead is used to carry OBC overhead, bandwidth map configuration information, and bandwidth report information for one BTC.
  • Burst Transfer Channel (BTC) overhead includes length field, sink node ID, up/down path identification, bandwidth map, bandwidth report, and other burst transfer channel related overhead.
  • the length field is composed of the number of OBs, the number of wavelengths, the number of bandwidth reports, and the CRC check; wherein the number of OBs and the number of wavelengths respectively represent the total number of OB slots and the number of wavelengths allocated to the BTC in the bandwidth map, thereby determining the bandwidth map length;
  • the number of bandwidth reports indicates the total number of bandwidth reports carried by this BTC, which can determine the bandwidth report length.
  • the sink node ID is used to indicate the sink node of this BTC.
  • the road sign and the down road sign respectively indicate whether the BTC is on the road and the down road in this control frame period.
  • the combination of the sink ID, the add-on identifier, and the drop identifier in the BTC with the source node ID in the BMS can uniquely identify a BTC.
  • the bandwidth map is used to carry the wavelength and OB slot configuration information of each BTC.
  • a bandwidth map consists of one or more submaps, each of which contains a wavelength field and an OB slot assignment field.
  • the wavelength field consists of the wavelength ID and the number of OBs, where the wavelength ID uniquely identifies the assigned wavelength number and the number of OBs represents the number of OBs allocated at this wavelength.
  • the OB slot allocation field uses bit mapping. Each bit corresponds to an OB slot position in the data frame. If the corresponding OB slot is assigned to this BTC, it is set to 1, otherwise it is set to 0.
  • control frame bandwidth map object coding structure uses the burst transmission multiplexing segment + the burst transmission channel to organize the two layers, which can be applied to all three OBTN nodes. Structure; (2) The wavelength and time slot configuration information in the bandwidth map uses bit mapping, which can effectively reduce the control frame length and reduce the control frame overhead.
  • FIG. 11 is a schematic diagram of an apparatus for updating a bandwidth map according to an embodiment of the present invention. As shown in FIG. 11, the apparatus of this embodiment includes:
  • the initialization module is configured to receive a bandwidth report reported by the control frame, create a new bandwidth map, and newly create a corresponding resource status table, and set all resource states in the newly created table to be available;
  • An allocation module is configured to transmit a channel uplink allocation structure across the primary node according to the bandwidth map to be updated, add a newly created bandwidth map across the primary node transmission channel downlink allocation structure, and trigger an update module;
  • the generating module is configured to allocate a wavelength and an optical burst slot one by one according to the bandwidth report reported by the control frame, add the newly created bandwidth map, generate a new bandwidth map, and trigger The update module;
  • a distribution module configured to carry the hop-by-hop distribution to the slave node by using the control frame;
  • the update module is configured to update the corresponding resource state table after being triggered.
  • the distribution module adds a newly created bandwidth map to the downlink distribution structure of the transmission channel of the primary node according to the bandwidth allocation map of the bandwidth map to be updated, and includes: traversing the bandwidth map to be updated, and The wavelength of the uplink allocation structure of the master node transmission channel and the position of the optical burst time slot are configured as the wavelength of the same transmission channel downlink allocation structure and the optical burst time slot position in the newly created bandwidth map.
  • the initializing module the newly created corresponding resource status table includes: a new source/sink resource status table of the current allocation period, a link resource status table of the current allocation period, and a sink resource status table of the next allocation period, the source / The sink resource status table is a three-dimensional matrix table, indicating an available state of the i-th transmitter/receiver of the node n at the mth optical burst slot position; the link resource status table is a three-dimensional matrix table, indicating a link The usable state of the wavelength w in the position of the mth optical burst slot.
  • the update module updating, by the update module, the corresponding resource status table after being triggered, including: updating a source/sink resource status table of the current allocation period and a link of the current allocation period for non-cross-node service a resource status table, a source resource status table of the current allocation period, a link resource status table of the current allocation period, and a sink resource status table of the next allocation period for the cross-node service;
  • the node downlink service updates the sink resource status table of the current allocation period and the link resource status table of the current allocation period.
  • the generating module, the wavelength resource and the optical burst time slot are allocated one by one for the current bandwidth request, including: querying the source resource state table and the sink resource state table according to the source node identifier and the sink node identifier of the current bandwidth request, for example, If the source resource status table and the sink resource status table have the same optical burst time slot, the routing table is queried according to the source node identifier and the sink node identifier of the current bandwidth request, if the routing table has Selecting a route, selecting a route, and querying the link resource status table according to the selected route according to the wavelength order. If the link has the same optical burst time slot resource at the same wavelength and the source end, select one. The wavelengths and optical burst slot positions that meet the requirements are allocated.
  • the generating module generates a new bandwidth map carrying wavelengths and optical burst slot configuration information of each burst transmission channel, where the new bandwidth map is composed of one or more submaps, each sub
  • the map includes a wavelength field and an optical burst slot allocation field, where the wavelength field is composed of a wavelength identifier and a number of optical bursts, and the optical burst slot allocation field is in a bit mapping manner, and each bit corresponds to a data frame.
  • An optical burst slot position is
  • the distribution module is configured to perform object coding according to the new bandwidth map, and the new bandwidth map carried by the control frame is hop-by-hop distributed to the slave node.
  • the generation module generates a complete new bandwidth map based on the wavelength and time slot assignment results, and the initialized new bandwidth map. Since the state table already shows the resource occupancy across the primary node during the wavelength and OB time slot allocation process, there is no resource conflict in the new bandwidth map.
  • the embodiment of the present invention further provides a computer program, including program instructions, when the program instruction is executed by the master node, so that the master node can perform the method of the embodiment of the present invention.
  • Embodiments of the present invention also provide a carrier carrying the above computer program.
  • all or a portion of the above steps may be accomplished by a program instructing the associated hardware, such as a read-only memory, a magnetic disk, or an optical disk.
  • all or part of the steps of the above embodiments may also be implemented using one or more integrated circuits (using an FPGA or an ASIC chip).
  • each module/unit in the above embodiment may be implemented in the form of hardware or in the form of a software function module. The invention is not limited to any specific form of combination of hardware and software.
  • the embodiments of the present invention can eliminate resource conflicts in the bandwidth map calculation and update process, and implement the dynamic allocation of bandwidth between any nodes, while reducing the packet loss rate and improving the network service carrying capacity.

Landscapes

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

Abstract

一种带宽地图更新的方法及装置,该方法包括:主节点收到控制帧携带的带宽报告后,新建带宽地图,新建资源状态表,将新建的资源状态表中所有资源状态置为可用;根据待更新的带宽地图的跨主节点传送通道上路分配结构,添加新建的带宽地图跨主节点传送通道下路分配结构,并按照其资源占用更新所述资源状态表;根据控制帧携带的带宽报告,为当前带宽请求逐一分配波长和光突发时隙,加入所述新建的带宽地图,生成新带宽地图,并更新所述资源状态表;通过控制帧携带所述新带宽地图逐跳分发至从节点。

Description

一种带宽地图更新的方法及装置 技术领域
本发明涉及光网络技术领域, 更具体的说, 涉及一种子波长全光交换网 络, 即光突发传送网 (Optical Burst Transport Network, 简称 OBTN ) 的波长 时隙分配和带宽地图更新方法和装置。
背景技术
全球数据流量爆炸式增长, 以视频和流媒体业务为代表的新兴业务快速 发展, 使动态、 高带宽和高质量要求的数据业务成为网络流量主体, 并驱动 网络向分组化演进。 在传送网方面, 可以看到, 从传统的 SDH ( Synchronous Digital Hierarchy, 同步数字体系) 电路交换网络, 发展到具备多业务接入功 能的 MSTP ( Multi-Service Transfer Platform,基于 SDH 的多业务传送平台), 并逐步演进至今天的 PTN ( Packet Transport Network, 分组传送网), 正是网 络流量数据化发展的结果。 究其根本, 电路交换网络仅能提供刚性的管道和 粗粒度交换, 无法有效满足数据业务的动态性和突发性需求, 而分组交换网 络的柔性管道和统计复用特性, 是天然适应于数据业务的。 然而, 目前的分 组交换基本上是基于电层处理的, 成本高, 能耗大, 随着流量的快速增长, 其处理瓶颈日渐凸显, 难以适应未来网络高速、 灵活、 低成本和低能耗的需 要。 光网络具备低成本、 低能耗和高速大容量的优势, 但传统的光电路交换 网络(如 WDM ( Wavelength Division Multiplexing,波分复用)和 OTN ( Optical Transport Network, 光传送网)仅能提供大粒度的刚性管道, 缺乏电分组交换 的灵活性, 不能有效的承载数据业务。
在接入网中, GPON ( Gigabit-Capable Passive Optical Network, 吉比特无 源光网络)技术一定程度上结合了光层和电层的优势。 在下行方向, 其釆用 光层广播的方式, 将 OLT ( Optical Line Terminal, 光线路终端)发送的下行 信号通过光分路器分发给各 ONU ( Optical Network Unit, 光网络单元) , 同 时,在下行帧头中携带上行帧的带宽地图, 以指示各 ONU上行数据的发送时 间和长度; 在上行方向, 各 ONU按照带宽地图指示发送数据, 经过光耦合器 复用至一条波长通道并上传至 OLT。 这样, GPON—方面具备光层高速大容 量和低成本的特点, 另一方面, 在上行方向上实现了多路数据的光层统计复 用, 提高了灵活性和带宽利用率。 GPON—般采用星形 /树形组网拓朴, 其工 作原理适合承载多点对单点的汇聚型流量(南北流量占主导地位) , 因此在 接入网中获得成功应用和大规模部署。
然而,针对非汇聚型应用场景, 如城域核心网和数据中心内部交换网络, 东西向流量占比很大, 甚至居于主导地位, GPON技术显然是不适合的 (东 西向流量需要 OLT电层转发, 且 GPON容量有限)。 光突发传送网 (Optical Burst Transport Network, 简称 OBTN )釆用基于 OB ( Optical Burst, 光突发) 的全光交换技术,具备网络任意节点对间光层带宽按需提供和快速调度能力 , 可实现对各种流量(如南北向突发流量、 东西向突发流量等)场景的动态适 应和良好支持, 能够提升资源利用效率和网络灵活性, 同时保留光层高速大 容量和低成本的优点, 且适用于星形 /树形 /环形各种网络拓朴。 图 1为 4节点 OBTN单向环形网络示意图, 每个节点配置一对快速可调谐突发发射机和快 速可调谐突发接收机(可扩展为多个) , 整网有两个波长作为数据通道, 一 个波长作为控制通道, 节点 A为主节点。 OBTN技术特征简述如下:
( 1 )数据通道中最基本的传输单元为 OB。 OB间有保护时间作为间隔, 一个或多个 OB组成一个数据帧,不同波长通道的相应 OB帧和 OB时隙起始 位置需要对齐。 数据通道釆用突发光接收机 /发射机, 突发数据在源宿节点对 间是光层直传, 不需要中间节点作电层转发。 源端需要将客户侧数据包汇聚 并封装至 OB发送。
( 2 )控制通道与数据通道相分离。 OBTN釆用独立的波长通道承载控制 信息, 其中包括 OAM ( Operations Administration and Maintenance , 操作维护 管理)信息、 用于搜集各节点带宽请求的带宽报告和指示各节点发送 /接收数 据的带宽地图, 且控制帧先于对应的数据帧发送。 控制通道可以釆用普通的 光接收机 /发射机作为收发设备, 在每个节点都进行电域处理, 以接收和更新 相应控制信息。 控制帧与数据帧的时序关系如图 2所示。
( 3 )釆用快速可调谐光器件实现基于 ΟΒ的全光交换。 OBTN节点可以 快速调节(ns (纳秒)级)发射机 /接收机的发射 /接收波长, 以根据带宽地图 选择相应的波长和 OB时隙进行突发数据发送 /接收, 以实现基于 OB的全光 交换。
( 4 )流量感知的实时光层资源调度。 OBTN采用集中式控制方式, 各从 节点通过控制帧周期性上报带宽请求至主节点, 主节点根据当前资源状态和 带宽分配策略进行波长和 OB时隙分配, 并将分配结果记入带宽地图, 再由 控制帧分发至各从节点, 以根据流量需求实现光层资源快速调度。
然而, 由于突发数据包在源宿节点对间光层直传, 而不进行电处理, 因 此受到波长一致性和时隙一致性的约束。 如图 1 , 节点 A发往节点 D的一个 突发数据包, 在节点 A由带宽地图指定在波长 λ 1相应数据帧的第 3个 ΟΒ 时隙上路, 由于光层直传, 且没有波长变换器和光緩存(波长变换器成本高 昂, 对信号质量有严重影响; 光緩存技术未突破) , 在经过节点 Β、 节点 C 至节点 D下路时,此突发数据包必须也占用波长 λ 1相应数据帧的第 3个 ΟΒ 时隙, 而不能更改波长和时隙位置。 由于波长时隙分配受到上述多重约束, 且带宽资源有限, 因此, 如分配不当, 会导致资源冲突, 引起大量丟包, 严 重降低网络性能。 ΟΒΤΝ中资源冲突主要包括下列三种:
( 1 )发射机资源冲突。 一个源端发射机于任意时隙位置能且仅能在一个 波长上发送突发数据。 如图, 节点 Α在波长 λ ΐ相应数据帧的第 3个 ΟΒ时 隙发送一个 A-->D的突发数据包; 此时, 如果带宽地图中还有一个以 A节点 为源节点的业务在波长 λ 2相应数据帧的第 3个 ΟΒ时隙上,则产生发射机资 源冲突。
( 2 )接收机资源冲突。 一个目的端接收机于任意时隙位置能且仅能在一 个波长上接收突发数据。 如图, 节点 D在波长 λ ΐ相应数据帧的第 3个 ΟΒ 时隙接收一个 -->0的突发数据包; 此时, 如果带宽地图中还有一个以 D节 点为目的节点的业务在波长 λ 2相应数据帧的第 3个 ΟΒ时隙上,则产生接收 机资源冲突。
( 3 )链路资源冲突。 同一链路上任意波长相应数据帧的任意时隙能且仅 能被分配一次。 如图, 业务 A-->D占用波长 λ 1相应数据帧的第 3个 ΟΒ时 隙; 此时, 如果带宽地图中还有业务 Β-->Α 占用波长 λ ΐ相应数据帧的第 3 个 ΟΒ时隙, 则在链路 BC和链路 CD上将产生链路资源冲突。 发明内容
本发明实施例提供一种带宽地图更新的方法及装置, 以在带宽地图计算 和更新过程中消除资源冲突。
本发明实施例提供的一种带宽地图更新的方法, 包括:
主节点收到控制帧携带的带宽报告后, 新建带宽地图, 新建资源状态表, 将新建的资源状态表中所有资源状态置为可用;
根据待更新的带宽地图的跨主节点传送通道上路分配结构, 添加新建的 带宽地图跨主节点传送通道下路分配结构, 并按照其资源占用更新所述资源 状态表;
根据控制帧携带的带宽报告, 为当前带宽请求逐一分配波长和光突发时 隙, 加入所述新建的带宽地图, 生成新带宽地图, 并更新所述资源状态表; 通过控制帧携带所述新带宽地图逐跳分发至从节点。
较佳地, 上述方法还具有下面特点: 所述根据待更新的带宽地图的跨主 节点传送通道上路分配结构, 添加新建的带宽地图跨主节点传送通道下路分 配结构, 包括:
遍历待更新的带宽地图, 将其中跨所述主节点传送通道上路分配结构的 波长和光突发时隙位置, 配置为所述新建的带宽地图中同一传送通道下路分 配结构的波长和光突发时隙位置。
较佳地, 上述方法还具有下面特点: 所述新建资源状态表包括: 新建当前分配周期的源 /宿资源状态表, 当前分配周期的链路资源状态表 和下一分配周期的宿资源状态表。
较佳地, 上述方法还具有下面特点:
所述源 /宿资源状态表为 N l M三维矩阵表,表示节点 n第 i个发射机 /接收机在第 m个光突发时隙位置的可用状态, n=l , 2, N; i=l , 2, ... , I; m=l , 2, M; N, I, M均为大于 1的整数, N表示 OBTN环网节点 数, I表示节点数据通道收发机数量, M表示一个数据帧的光突发时隙总数; 所述链路资源状态表为 L X W X M三维矩阵表, 表示链路 1中波长 w在 第 m个光突发时隙位置的可用状态, 1=1 , 2, ... ... , L; w=l , 2, W; m=l , 2, M; L, W, M均为大于 1的整数, L表示链路数, W表示数 据通道波长数, M表示一个数据帧的光突发时隙总数。
较佳地, 上述方法还具有下面特点: 所述更新相应状态表包括: 所述更新资源状态表包括:
对于非跨主节点业务, 更新所述当前分配周期的源 /宿资源状态表和所述 当前分配周期的链路资源状态表;
对于跨主节点上路业务, 更新所述当前分配周期的源资源状态表、 所述 当前的链路资源状态表和所述下一分配周期的宿资源状态表; 对于跨主节点 下路业务, 更新所述当前分配周期的宿资源状态表和所述当前分配周期的链 路资源状态表。
较佳地, 上述方法还具有下面特点: 所述为当前带宽请求逐一分配波长 和光突发时隙, 包括:
根据所述当前带宽请求的源节点标识和宿节点标识, 依次查询源资源状 态表和宿资源状态表, 如所述源资源状态表和所述宿资源状态表有一致的光 突发时隙可用, 则根据所述当前带宽请求的源节点标识和宿节点标识查询路 由表, 如所述路由表中有未选择的路由, 则选取一路由, 根据所选路由依波 长次序查询所述链路资源状态表, 如链路在同一波长和源端宿端有一致的光 突发时隙资源可用,则选取一个满足要求的波长和光突发时隙位置进行分配。
较佳地, 上述方法还具有下面特点:
所述新带宽地图携带每个突发传送通道的波长和光突发时隙配置信息, 所述新带宽地图由一个或多个子地图组成, 每个子地图包含一个波长字段和 一个光突发时隙分配字段。
较佳地, 上述方法还具有下面特点:
所述波长字段由波长标识和光突发数量组成, 所述光突发时隙分配字段 釆用比特映射方式, 每个比特对应数据帧中的一个光突发时隙位置。
较佳地, 上述方法还具有下面特点: 所述通过控制帧携带所述新带宽地 图逐跳分发至从节点, 包括:
根据所述新带宽地图进行对象编码, 由控制帧携带编码后的新带宽地图 逐跳分发至从节点。
为了解决上述问题, 本发明实施例还提供了一种带宽地图更新的装置, 其中, 初始化模块、 分配模块、 生成模块、 分发模块和更新模块, 其中: 所述初始化模块设置为: 收到控制帧携带的带宽^艮告后, 新建带宽地图, 新建资源状态表, 将新建的资源状态表中所有资源状态置为可用;
所述分配模块设置为: 根据待更新的带宽地图的跨主节点传送通道上路 分配结构, 添加新建的带宽地图跨主节点传送通道下路分配结构, 并触发所 述更新模块;
所述生成模块设置为: 根据控制帧上报的带宽报告, 为当前带宽请求逐 一分配波长和光突发时隙, 加入所述新建的带宽地图, 生成新带宽地图, 并 触发所述更新模块;
所述分发模块, 其设置为: 通过控制帧携带所述新带宽地图逐跳分发至 从节点; 以及
所述更新模块, 其设置为: 受触发后更新所述资源状态表。
较佳地, 上述装置还具有下面特点:
所述分配模块是设置为以如下方式根据待更新的带宽地图的跨主节点传 送通道上路分配结构,添加新建的带宽地图跨主节点传送通道下路分配结构: 遍历待更新的带宽地图, 将其中跨所述主节点传送通道上路分配结构的波长 和光突发时隙位置, 配置为所述新建的带宽地图中同一传送通道下路分配结 构的波长和光突发时隙位置。
较佳地, 上述装置还具有下面特点:
所述初始化模块是设置为以如下方式新建资源状态表包括: 新建当前分 配周期的源 /宿资源状态表和当前分配周期的链路资源状态表和下一分配周 期的宿资源状态表, 所述源 /宿资源状态表为 N X I M三维矩阵表, 表示节 点 n第 i个发射机 /接收机在第 m个光突发时隙位置的可用状态, n=l , 2, N; i=l , 2, I; m=l , 2, M; N, I, M均为大于 1 的整数, N表 示 OBTN环网节点数, I表示节点数据通道收发机数量, M表示一个数据帧 的光突发时隙总数; 所述链路资源状态表为 L X W X M三维矩阵表, 表示链 路 1中波长 w在第 m个光突发时隙位置的可用状态, 1=1 , 2, ... ... , L; w=l ,
2, ..· , W; m=l , 2, M; L, W, M均为大于 1的整数, L表示链路数, W表示数据通道波长数, M表示一个数据帧的光突发时隙总数。
较佳地, 上述装置还具有下面特点:
所述更新模块是设置为以如下方式更新所述相应的资源状态表: 对于非 跨主节点业务, 更新所述当前分配周期的源 /宿资源状态表和所述当前分配周 期的链路资源状态表; 对于跨主节点上路业务, 更新所述当前分配周期的源 资源状态表、 所述当前分配周期的链路资源状态表和所述下一分配周期的宿 资源状态表; 对于跨主节点下路业务, 更新所述当前分配周期的宿资源状态 表和所述当前分配周期的链路资源状态表。
较佳地, 上述装置还具有下面特点:
所述生成模块是设置为以如下方式为当前带宽请求逐一分配波长和光突 发时隙: 根据所述当前带宽请求的源节点标识和宿节点标识, 依次查询源资 源状态表和宿资源状态表, 如所述源资源状态表和所述宿资源状态表有一致 的光突发时隙可用, 则根据所述当前带宽请求的源节点标识和宿节点标识查 询路由表, 如所述路由表中有未选择的路由, 则选取一路由, 根据所选路由 依波长次序查询所述链路资源状态表, 如链路在同一波长和源端宿端有一致 的光突发时隙资源可用, 则选取一个满足要求的波长和光突发时隙位置进行 分配。
较佳地, 上述装置还具有下面特点:
所述生成模块 , 生成的新带宽地图携带每个突发传送通道的波长和光突 发时隙配置信息, 所述新带宽地图由一个或多个子地图组成, 每个子地图包 含一个波长字段和一个光突发时隙分配字段, 所述波长字段由波长标识和光 突发数量组成, 所述光突发时隙分配字段釆用比特映射方式, 每个比特对应 数据帧中的一个光突发时隙位置。
较佳地, 上述装置还具有下面特点: 所述分发模块, 是设置为根据所述新带宽地图进行对象编码, 由控制帧 携带编码后的新带宽地图逐跳分发至从节点。
本发明实施例还提供一种计算机程序, 包括程序指令, 当该程序指令被 主节点执行时, 使得该主节点可执行上述方法。
本发明实施例还提供一种载有上述计算机程序的载体。
综上, 本发明实施例提供一种带宽地图更新的方法及装置, 在 OBTN网 络中, 在带宽地图计算和更新过程中消除资源冲突, 在实现任意节点对间带 宽动态分配的同时, 可以降低丟包率, 提高网络业务承载能力。 附图概述
图 1是 OBTN环网基本结构的示意图;
图 2是 OBTN环网中控制帧与数据帧相对时序的示意图;
图 3是本发明实施例的源 /宿资源状态表举例的示意图;
图 4是本发明实施例的链路资源状态表举例的示意图;
图 5是本发明实施例的路由表举例的示意图;
图 6是本发明实施例的带宽地图更新方法的流程图;
图 7是本发明实施例的波长和时隙分配方法流程图;
图 8是本发明实施例的带宽地图对象编码示意图;
图 9a至图 9d是本发明实施例的带宽地图更新的示意图;
图 10a为针对图 1网络场景的 A-->C的波长和 OB时隙分配的示例图; 图 10b为针对图 1网络场景的 C-->B的波长和 OB时隙分配的示例图; 图 11为本发明实施例的带宽地图更新的装置的示意图。 本发明的较佳实施方式
OBTN尽管具备全光子波长交换的优势, 但由于光緩存技术和波长变换 技术的不成熟, 使其无法借鉴使用传统电层分组设备的 "存储-转发" 方式, 使得资源冲突可能性极大增加, 可能造成大量丟包。 为此, 有必要在带宽地 图计算阶段考虑资源冲突问题, 通过合理的波长时隙资源分配, 并结合相应 的带宽地图更新方法, 以避免资源冲突, 提升 OBTN网络效率; 同时, 由于 OBTN网络需要实现快速资源调度, 因此波长时隙分配算法复杂度必须合理, 以满足实时性需求。 然而, 当前 OBTN相关的文献和专利中, 都没有考虑此 问题。
下文中将结合附图对本发明的实施例进行详细说明。 需要说明的是, 在 不冲突的情况下, 本申请中的实施例及实施例中的特征可以相互任意组合。
本发明实施例对 OBTN组网场景的要求: 支持环形拓朴组网, 可以是单 纤单向, 也可以是双纤双向或者多纤双向。
本发明实施例对 OBTN节点结构的要求: 网络中的节点根据流量需求, 可配置一对或多对数据通道收发机, 数据通道收发机可釆用 3种方式: a. 快 速可调谐突发发射机( Fast Tunable Burst Mode Transmitter, FT-BMT ) +普通 接收机(Fast Tunable Burst Mode Receiver, FT-BMR ) ; b. 普通发射机 + FT-BMR; c. FT-BMT+FT-B 。
图 1 为 OBTN单向环网基本结构图, 其网络场景如下:
1 ) 网络拓朴为 4节点单向环网, 节点 A为主节点, 其他节点为从节点;
2 )各节点在数据通道配置一对快速可调谐突发发射机 /接收机, 在控制 通道配置一对普通发射机 /接收机;
3 )环网数据通道配置 2个波长, 控制通道配置一个波长;
4 )数据帧包含 10个 OB时隙, 环网数据环长为数据帧长度的 4倍;
5 )控制帧由主节点产生, 由从节点更新,沿环传送一周并于主节点终结;
6 ) DBA (动态带宽分配)算法一个数据帧执行一次。
为了详细说明本发明内容, 主要针对图 1场景进行实施例描述, 但除此 场景外, 本发明还适合任何单向 /双向 OBTN环网场景, 而星形和树形拓朴较 为简单, 对发明实施例的内容进行简化即可适用。
图 1中带宽地图更新主要分为以下四个阶段: 1 ) 带宽请求上报:
由控制帧携带带宽报告至主节点,主节点接收并将其量化为各节点对 OB 时隙请求;
2 ) 带宽地图分配:
由主节点根据当前资源状态和带宽请求,执行 DBA算法, 进行波长和时 隙分配, 生成新带宽地图;
3 ) 带宽地图分发:
由控制帧携带带宽地图, 提前于相应数据帧逐跳发送至从节点, 从节点 接收并生成本节点上路地图和下路地图, 同时, 控制帧在每个从节点搜集带 宽请求;
4 ) 带宽地图更新与执行:
从节点收到控制帧后, 按照控制帧和数据帧的相对延时, 延后一段时间 再令新带宽地图生效。
图 9a至图 9d为针对图 1网络场景的带宽地图更新的实施例, 详细流程 如下:
当主节点收到控制帧的带宽报告, 进入带宽地图计算过程时, 首先执行 带宽地图更新, 如图 6所示, 步骤如下:
步骤 101 : 主节点收到控制帧上报的带宽报告后, 新建带宽地图, 新建 相应的资源状态表, 将新建的表中所有资源状态置为可用;
主节点新建一张未分配带宽地图, 称为新带宽地图。 新建当前分配周期 的源 /宿资源状态表和链路资源状态表; 新建下一分配周期宿资源状态表, 简 称下周期宿资源状态表; 所有资源状态置为可用。
在主节点收到控制帧带宽报告, 进入带宽地图计算过程, 第一步初始化 情况下新建新带宽地图和上述三个表。
本实施例中, 如图 9a所示, 根据网络配置, 包含 2个波长和 10个 OB 时隙; 初始化当前分配周期的源 /宿资源状态表、 下周期宿资源状态表。 本实 施例中, 下周期宿资源状态表代表本带宽地图对应的当前数据帧之后第 4个 数据帧的宿资源状态表(因为当前带宽地图中跨主节点上路业务分配的 OB 时隙会穿通主节点,并被看作下周期带宽地图的跨主节点下路业务分配的 OB 时隙, 占用下周期数据帧对应的宿下路资源)和链路资源状态表, 并将其全 部置 1 , 表示资源未被占用。
步骤 102: 根据待更新带宽地图的跨主节点传送通道上路分配结构, 添 加新带宽地图跨主节点传送通道下路分配结构, 并按照其资源占用, 更新当 前分配周期的宿资源状态表和链路资源状态表; 主节点遍历待更新带宽地图, 即旧带宽地图, 将其中跨主节点传送通道 上路分配结构的波长和 OB时隙位置, 配置为新带宽地图中同一传送通道下 路分配结构的波长和时隙位置, 同时更新当前分配周期的宿资源状态表和当 前分配周期的链路资源状态表。 旧带宽地图的跨主节点传送通道上路业务所分配的 OB时隙, 在穿通主 节点后, 会占用新带宽地图生效周期内的相应链路和宿下路资源, 即被看作 是新带宽地图的跨主节点传送通道下路业务所分配的 OB时隙, 所以, 需要 遍历旧带宽地图, "将其中跨主节点传送通道上路分配结构的波长和 OB时 隙位置, 配置为新带宽地图中同一传送通道下路分配结构的波长和时隙位 置,, 。
待更新带宽地图中, 跨主节点传送通道上路分配结构包括: 波长 1 第 5 个 OB时隙的 D-->C上路, 波长 2第 2个 OB时隙的 C-->B上路, 第 3个 OB 时隙 D-->B上路, 第 4个 OB时隙 D-->C上路, 第 6个 OB时隙 D-->C上路。
将其转化为新带宽地图的跨主节点传送通道下路分配结构, 包括: 波长 1第 5个 OB时隙的 D-->C下路, 波长 2第 2个 OB时隙的 C-->B下路, 第 3 个 OB时隙 D-->B下路, 第 4个 OB时隙 D-->C下路, 第 6个 OB时隙 D-->C 下路。
同时, 按照跨主节点传送通道下路分配结构的资源占用状态, 更新当前 分配周期的宿资源状态表和链路资源状态表。 如波长 1 第 5 个 OB 时隙的 D-->C下路, 需占用第 5个 OB时隙的 C节点下路资源, 及链路 AB和 BC在 波长 1第 5个 OB时隙的时隙资源, 因此将其置 0标识不可用, 其后在波长 和时隙分配过程中则不会选择此资源。 最终资源状态表更新结果如图 9b 所 本实施例中, 待更新带宽地图代表当前数据帧之前的第 4个数据帧对应 的带宽地图, 因为待更新带宽地图的跨主节点上路业务分配的 OB时隙会穿 通主节点, 并被看作当前带宽地图的跨主节点下路业务分配的 OB时隙, 占 用本数据帧对应的宿下路资源。 因为前后周期带宽地图间存在相互影响, 因 此, 在本实施例中, 主节点需要至少保存 4个带宽地图。
步骤 103: 根据控制帧上报的带宽报告, 为当前带宽请求逐一分配波长 和 OB时隙, 加入新带宽地图, 并更新相应状态表。
对于非跨主节点业务, 需要更新当前分配周期的源 /宿资源状态表和链路 资源状态表; 对于跨主节点上路业务, 需要更新当前分配周期的源资源状态 表、 链路资源状态表和下周期宿资源状态表; 对于跨主节点下路业务, 更新 所述当前的宿资源状态表和所述当前分配周期的链路资源状态表。
本实施例中, 对带宽请求进行量化及波长和 OB时隙分配, 并更新相应 资源状态表。
如 A-->C请求分配 1个 OB时隙, 经过状态表查询和波长时隙分配算法 后, 为其分配波长 1的第 2个 OB时隙, 其资源占用如下: 节点 A第 2个 OB 时隙的上路资源,节点 C第 2个 OB时隙的下路资源,及链路 AB和 BC在波 长 1第 2个 OB时隙的时隙资源。在资源表中将相应位置置为 0, 以避免重复 使用造成冲突。
又如, C~>B上路请求分配 1个 OB时隙 ,经过状态表查询和波长时隙分 配算法后, 为其分配波长 2的第 3个 OB时隙, 其资源占用如下: 节点 c第 3 个 OB时隙的上路资源, 节点 B第 3个 OB时隙在下一周期的下路资源, 及 链路 CD和 DA在波长 2第 3个 OB时隙的时隙资源。在资源表中将相应位置 置位 0。 注意, C-->B的路由穿通主节点, 因此只占用 C节点的当前分配周期 的上路资源和链路 CD和 DA的时隙资源, 而 B节点下路资源和链路 AB时 隙资源的占用在下一分配周期带宽地图中体现。最终带宽地图和资源状态表, 如图 9c所示。
步骤 104: 主节点通过控制帧携带所述新带宽地图逐跳分发至从节点。 完成新带宽地图构建, 在主节点保存, 并按照带宽地图编码方式生成带 宽地图对象, 由控制帧携带逐跳分发至从节点。
带宽地图需要携带在控制帧中传递至从节点, 以指导从节点在哪些波长 和时隙上下路哪些业务。 因此, 带宽地图需要有一致且标准化的编码方式, 以方便主节点编码, 控制帧携带和从节点识别。
本实施例中, 对带宽地图对象进行编码, 并与控制帧中携带至从节点, 控制帧与数据帧时序如图 2所示; 从节点根据带宽地图生成发送波长时隙表 和接收波长时隙表, 如图 1所示, 经过一定时延(节点时延通过测量确定, 并在控制帧中携带, 其实现机制不在本专利讨论范围内)生效, 以控制节点 的数据发送和接收。
如 D-->C下路在本带宽地图共占用 3个 OB时隙, 分别是波长 1的第 5 个时隙, 在相应 BTC (突发传送通道)开销的 OB 时隙分配编码中体现为 0x0010 (第 5个 bit置位 ) , 波长 2的第 4 和第 6个时隙, 在 OB时隙分配编 码中体现为 0x0028 (第 4和第 6个 bit置位) 。 由于 D-->C下路在本带宽地 图中属于跨越主节点传送通道下路分配结构, 在本带宽地图生效范围内仅接 收, 因此上路位置 0, 下路位置 1 ; 跨越主节点传送通道下路分配结构不携带 带宽报告; 此 BTC的源节点在所属 BMS (突发复用段)开销中指明。
如 A-->C在本带宽地图中属于非跨主节点传送通道, 在本带宽地图生效 范围内发送和接收, 因此上路位和下路位都置 1 ; 以主节点为源节点的 BTC 开销中不需要携带带宽报告, 其可由本地读取直接获得。
又如 C-->B上路在本带宽地图中属于跨主节点传送通道上路分配结构, 在本带宽地图生效范围内仅发送, 因此仅上路位置 1 ; 其可能携带带宽报告, 但主节点生成的带宽地图不包括带宽报告, 而是由控制帧携带其到达从节点 时, 从节点根据需求在控制帧相应的 BTC开销中添加并上报, 如图 9d所示。
本发明实施例中的需要使用如下状态表: 源 /宿资源状态表、 链路资源状 态表、 路由表:
源 /宿资源状态表为 X I 三维矩阵, 其中, N表示 OBTN环网节点 数, I表示节点数据通道收发机数量, M表示一个数据帧的 OB时隙总数。 其 表示节点 n第 i个发射机 /接收机在第 m个 OB时隙位置的可用状态, 1为资 源可用, 0为资源占用。 N, I, M均为大于 1的整数, n=l, 2, N; i=l, 2, I; m=l, 2, M。
链路资源状态表为 L W X M三维矩阵, 其中, L表示链路数, W表示 数据通道波长数, M表示一个数据帧的 OB时隙总数。其表示链路 1中波长 w 在第 m个 OB时隙位置的可用状态, 1为资源可用, 0为资源占用。 L, W, M均为大于 1的整数, 1=1, 2, ...... , L; w=l, 2, ..·, W; m=l, 2, ...,
Μ。
省空间, 并提高运算效率。
路由表为 N X N X PXL四维矩阵, 其中, N表示 OBTN环网节点数, P 表示节点对(s. d)的路由数, L表示链路数。 对于单向环网, 值为 1; 对于双 向环网, 值为 2。 其表示节点 s到节点 d的第 p条路由是否经过链路 1, 1为 经过, 反之为 0。 N, P, L均为大于 1的整数, n=l, 2, N; p=l, 2, ..., Ρ; 1=1, ..., L。
图 3、 4、 5分别是本实施例针对图 1场景的源 /宿资源状态表、 链路资源 状态表和路由表的示例。
本实施例中的带宽地图更新方法创新点和优势总结如下: ( 1 )通过资源 状态表, 能够有效标识跨主节点传送通道在本带宽地图周期的资源占用及当 前已分配传送通道的资源占用, 使得波长和时隙分配时不会产生资源冲突, 因此生成的带宽地图也不会出现资源冲突; (2)本带宽地图更新方法在波长 和时隙分配过程中考虑资源冲突问题, 能够有效提升网络承载能力, 资源利 用率高; 而现有带宽地图更新方法, 釆取带宽地图生成后检测冲突, 并撤销 带宽地图中冲突资源相关的已分配传送通道来避免冲突, 导致带宽分配不稳 定, 资源利用率低, 网络性能难以保证。
本实施例中基于状态表的波长和 OB时隙分配方法, 如图 7所示, 步骤 下: 步骤 201 : 从带宽请求集合中选取一条带宽请求;
具体选取算法和公平性保证机制不在本发明讨论范围内。
步骤 202: 根据所选取请求的源宿节点 ID, 依次查询源资源状态表和宿 资源状态表;
对于非跨主节点业务, 查询当前分配周期源宿资源状态表, 对于跨主节 点业务, 查询当前分配周期源资源状态表和下一分配周期宿资源状态表。 如 果源资源状态表和宿资源状态表有一致的 OB时隙可用,转至步骤 203; 如果 遍历状态表后,源宿资源状态表无一致时隙可用,则分配失败,转至步骤 205。
步骤 203: 根据所选取请求的源宿节点 ID查询路由表, 如果有未选择路 由, 选取此路由, 将其标记为已选择, 转至步骤 204; 如果已遍历所有路由, 而未完成分配, 则转至步骤 205。
步骤 204: 根据路由依波长次序查询链路资源状态表, 如果链路在同一 波长和源端宿端有一致的 OB时隙资源可用, 则选取一个满足要求的波长和 OB时隙位置进行分配, 将其添加至带宽地图, 分配完成; 如果遍历各波长都 没有一致的 OB时隙位置, 则转至步骤 203。
步骤 205: 从带宽请求集合去除此请求, 波长和 OB时隙分配失败。
图 10a为针对图 1网络场景的 A-->C的波长和 OB时隙分配的示例图, 图 10b为针对图 1网络场景的 C-->B的波长和 OB时隙分配的示例图。 这里 假设在图 9c的带宽地图更新执行完后还有 A-->C和 C-->B上路各请求 1个 OB时隙, 下面通过具体过程说明波长和时隙分配方法:
对于非跨主节点业务 A-->C, 如图 10a所示, 分配步骤如下:
( 1 )查找前 A节点当前分配周期源资源状态表和 C节点当前分配周期 宿资源状态表, 获取一致可用时隙集合: 时隙 7~10 (如釆用 bit映射方式, 可以简单的釆用与操作获得结果, 效率 4艮高, 能够保证实时性) ;
( 2 )查询路由表, 获取 A~>C的唯一路由(单向环网)穿通链路 AB和
BC;
( 3 )查找链路 AB和链路 BC的资源状态表, 获取源宿和链路各波长一 致可用时隙集合: 波长 1的 7~10时隙和波长 2的 7~10时隙;
(4)根据算法策略,从可用集合中选择一个波长和 OB时隙进行分配: 在 本实施例中, 分配波长 1的第 7个 OB时隙。 同时, 将其添加到带宽地图, 并更新 A节点当前分配周期源资源状态表、 C节点当前分配周期宿资源状态 表及链路 AB和 BC在波长 1的资源状态表。
对于跨主节点上路业务 C-->B, 如图 10b所示, 分配步骤如下:
( 1 )查找 C节点当前分配周期源资源状态表和 B节点下一分配周期宿 资源状态表, 获取一致可用时隙集合: 时隙 4、 时隙 6~10;
( 2 )查询路由表, 获取 C-->B的唯一路由 (单向环网) 穿通链路 CD、 DA和 AB;
( 3 )查找链路 CD和链路 DA的资源状态表(在本带宽地图的生效范围 内只占用 CD和 DA的链路资源,穿通主节点后 AB链路的资源占用体现在下 一周期 ) , 获取源宿和链路各波长一致可用时隙集合: 波长 1的 7~10时隙和 波长 2的 4、 6~10时隙;
( 4 )根据算法策略, 从可用集合中选择一个波长和 OB时隙进行分配: 在本实施例中, 分配波长 2的第 4个 OB时隙。 同时, 将其添加到带宽地图, 并更新 C节点当前分配周期源资源状态表、 B节点下一分配周期宿资源状态 表及链路 CD和 DA在波长 2的资源状态表。
本实施例中的波长和时隙分配方法的创新点和优势如下: ( 1 )资源状态 表可以釆取比特映射方式, 能够有效减少存储空间需求; (2 )基于状态表的 波长和时隙分配方法只需要简单的查表和与或操作, 复杂度低, 符合实时计 算和快速资源调度的需求, 且适于硬件实现; (3 )本实施例中波长和时隙分 配方法适合全部三种 OBTN节点结构; (4 )目前未有解决此问题的相关专利 和文献。
本实施例中的控制帧带宽地图对象编码结构, 如图 8所示。 本申请仅关 注带宽地图相关的编码结构, 对于控制帧其他对象结构不详细展开。
突发复用段(BMS )开销用于指定一个节点 (源节点) , 并携带与此节 点相关的 OAM开销。 BMS开销主要包括源节点 ID、 节点 OAM开销和长度 指示三部分, 其中长度指示由带宽地图有无标识、 突发传送通道(BTC )数 量和 CRC校验三部分组成。 BTC数量表示以 BMS中节点 ID为源节点的 BTC 总数, 而带宽地图标识指明这些 BTC是否携带带宽地图。 对于一个 n节点的 光突发环网, 控制帧中包括 n个突发复用段(BMS )开销。 突发传送通道(BTC )开销用于携带一个 BTC的 OAM开销、 带宽地图 配置信息和带宽报告信息。 突发传送通道(BTC )开销包括长度字段、 宿节 点 ID、 上 /下路标识、 带宽地图、 带宽报告和其他突发传送通道相关开销。 长 度字段由 OB数量、 波长数量、 带宽报告数量和 CRC校验组成; 其中 OB数 量和波长数量分别表示带宽地图中为此 BTC分配的 OB时隙总数和波长数, 以此可以确定带宽地图长度;带宽报告数量表示此 BTC携带的带宽报告总数, 以此可以确定带宽报告长度。 宿节点 ID用以指明此 BTC的宿节点。 上路标 识和下路标识分别表示此 BTC是否在此控制帧周期上路和下路。 BTC中的宿 节点 ID、上路标识和下路标识与 BMS中的源节点 ID组合可以唯一标识一个 BTC。
带宽地图用以携带每个 BTC的波长和 OB时隙配置信息。 一个带宽地图 由一个或多个子地图组成, 每个子地图包含一个波长字段和一个 OB时隙分 配字段。 波长字段由波长 ID和 OB数量组成 , 其中波长 ID唯一标识所分配 的波长编号, OB数量表示在此波长上所分配的 OB数量。 OB时隙分配字段 釆用比特映射方式, 每个比特对应数据帧中的一个 OB时隙位置, 如相应 OB 时隙分配给此 BTC则置 1 , 否则置 0。
本实施例中控制帧带宽地图对象编码结构的创新点和优势如下: ( 1 )控 制帧釆用突发传送复用段 +突发传送通道两层的组织方式,能够适用于全部三 种 OBTN节点结构; ( 2 )带宽地图中的波长和时隙配置信息釆用比特映射方 式, 能够有效减少控制帧长度, 降低控制帧开销。
图 11为本发明实施例的带宽地图更新的装置的示意图, 如图 11所示, 本实施例的装置包含:
初始化模块, 设置为收到控制帧上报的带宽报告后, 新建带宽地图, 新 建相应的资源状态表, 将新建的表中所有资源状态置为可用;
分配模块, 设置为根据待更新的带宽地图的跨主节点传送通道上路分配 结构, 添加新建的带宽地图跨主节点传送通道下路分配结构, 并触发更新模 块;
生成模块, 设置为根据控制帧上报的带宽报告, 为当前带宽请求逐一分 配波长和光突发时隙, 加入所述新建的带宽地图, 生成新带宽地图, 并触发 所述更新模块;
分发模块, 设置为通过控制帧携带所述新带宽地图逐跳分发至从节点; 所述更新模块, 设置为受触发后更新所述相应的资源状态表。
其中, 所述分配模块, 根据待更新的带宽地图的跨主节点传送通道上路 分配结构, 添加新建的带宽地图跨主节点传送通道下路分配结构, 包括: 遍 历待更新的带宽地图, 将其中跨所述主节点传送通道上路分配结构的波长和 光突发时隙位置, 配置为所述新建的带宽地图中同一传送通道下路分配结构 的波长和光突发时隙位置。
其中, 所述初始化模块, 新建相应的资源状态表包括: 新建当前分配周 期的源 /宿资源状态表和当前分配周期的链路资源状态表和下一分配周期的 宿资源状态表, 所述源 /宿资源状态表为三维矩阵表, 表示节点 n第 i个发射 机 /接收机在第 m个光突发时隙位置的可用状态;所述链路资源状态表为三维 矩阵表, 表示链路 1中波长 w在第 m个光突发时隙位置的可用状态。
其中, 所述更新模块, 受触发后更新所述相应的资源状态表, 包括: 对 于非跨主节点业务, 更新所述当前分配周期的源 /宿资源状态表和所述当前分 配周期的链路资源状态表; 对于跨主节点上路业务, 更新所述当前分配周期 的源资源状态表、 所述当前分配周期的链路资源状态表和所述下一分配周期 的宿资源状态表; 对于跨主节点下路业务, 更新所述当前分配周期的宿资源 状态表和所述当前分配周期的链路资源状态表。
其中, 所述生成模块, 为当前带宽请求逐一分配波长和光突发时隙, 包 括: 根据所述当前带宽请求的源节点标识和宿节点标识, 依次查询源资源状 态表和宿资源状态表, 如所述源资源状态表和所述宿资源状态表有一致的光 突发时隙可用, 则根据所述当前带宽请求的源节点标识和宿节点标识查询路 由表, 如所述路由表中有未选择的路由, 则选取一路由, 根据所选路由依波 长次序查询所述链路资源状态表, 如链路在同一波长和源端宿端有一致的光 突发时隙资源可用,则选取一个满足要求的波长和光突发时隙位置进行分配。
其中, 所述生成模块, 生成的新带宽地图携带每个突发传送通道的波长 和光突发时隙配置信息, 所述新带宽地图由一个或多个子地图组成, 每个子 地图包含一个波长字段和一个光突发时隙分配字段, 所述波长字段由波长标 识和光突发数量组成, 所述光突发时隙分配字段釆用比特映射方式, 每个比 特对应数据帧中的一个光突发时隙位置。
其中, 所述分发模块, 是设置为根据所述新带宽地图进行对象编码, 由 控制帧携带编码后的新带宽地图逐跳分发至从节点。
生成模块根据波长和时隙分配结果, 及初始化的新带宽地图, 生成完整 的新带宽地图。 由于在波长和 OB时隙分配过程中, 状态表已经体现了跨主 节点资源占用情况, 因此新带宽地图不存在资源冲突。
本发明实施例还提供一种计算机程序, 包括程序指令, 当该程序指令被 主节点执行时, 使得该主节点可执行本发明实施例的方法。
本发明实施例还提供一种载有上述计算机程序的载体。
本领域普通技术人员可以理解上述方法中的全部或部分步骤可通过程序 来指令相关硬件完成, 所述程序可以存储于计算机可读存储介质中, 如只读 存储器、 磁盘或光盘等。 可选地, 上述实施例的全部或部分步骤也可以使用 一个或多个集成电路(用 FPGA或 ASIC芯片 )来实现。 相应地, 上述实施 例中的各模块 /单元可以釆用硬件的形式实现, 也可以釆用软件功能模块的形 式实现。 本发明不限制于任何特定形式的硬件和软件的结合。
以上仅为本发明的优选实施例, 当然, 本发明还可有其他多种实施例, 在不背离本发明精神及其实质的情况下, 熟悉本领域的技术人员当可根据本 发明作出各种相应的改变和变形, 但这些相应的改变和变形都应属于本发明 所附的权利要求的保护范围。
工业实用性 本发明实施例可以在带宽地图计算和更新过程中消除资源冲突, 在实现 任意节点对间带宽动态分配的同时, 可降低丟包率, 提高网络业务承载能力。

Claims

权 利 要 求 书
1、 一种带宽地图更新的方法, 包括:
主节点收到控制帧携带的带宽报告后, 新建带宽地图, 新建资源状态表, 将新建的资源状态表中所有资源状态置为可用;
根据待更新的带宽地图的跨主节点传送通道上路分配结构, 添加新建的 带宽地图跨主节点传送通道下路分配结构, 并按照其资源占用更新所述资源 状态表;
根据控制帧携带的带宽报告, 为当前带宽请求逐一分配波长和光突发时 隙, 加入所述新建的带宽地图, 生成新带宽地图, 并更新所述资源状态表; 通过控制帧携带所述新带宽地图逐跳分发至从节点。
2、 如权利要求 1所述方法, 其中, 所述根据待更新的带宽地图的跨主节 点传送通道上路分配结构, 添加新建的带宽地图跨主节点传送通道下路分配 结构, 包括:
遍历待更新的带宽地图, 将其中跨所述主节点传送通道上路分配结构的 波长和光突发时隙位置, 配置为所述新建的带宽地图中同一传送通道下路分 配结构的波长和光突发时隙位置。
3、 如权利要求 1所述方法, 其中, 所述新建资源状态表包括:
新建当前分配周期的源 /宿资源状态表, 当前分配周期的链路资源状态表 和下一分配周期的宿资源状态表。
4、 如权利要求 3所述方法, 其中,
所述源 /宿资源状态表为 N l M三维矩阵表,表示节点 n第 i个发射机 /接收机在第 m个光突发时隙位置的可用状态, n=l , 2, N; i=l , 2, ... , I; m=l , 2, M; N, I, M均为大于 1的整数, N表示 OBTN环网节点 数, I表示节点数据通道收发机数量, M表示一个数据帧的光突发时隙总数; 所述链路资源状态表为 L X W X M三维矩阵表, 表示链路 1中波长 w在 第 m个光突发时隙位置的可用状态, 1=1 , 2, ... ... , L; w=l , 2, W; m=l , 2, M; L, W, M均为大于 1的整数, L表示链路数, W表示数 据通道波长数, M表示一个数据帧的光突发时隙总数。
5、 如权利要求 3所述方法, 其中, 所述更新资源状态表包括:
对于非跨主节点业务, 更新所述当前分配周期的源 /宿资源状态表和所述 当前分配周期的链路资源状态表;
对于跨主节点上路业务, 更新所述当前分配周期的源资源状态表、 所述 当前的链路资源状态表和所述下一分配周期的宿资源状态表; 对于跨主节点 下路业务, 更新所述当前分配周期的宿资源状态表和所述当前分配周期的链 路资源状态表。
6、 如权利要求 1所述方法, 其中, 所述为当前带宽请求逐一分配波长和 光突发时隙, 包括:
根据所述当前带宽请求的源节点标识和宿节点标识, 依次查询源资源状 态表和宿资源状态表, 如所述源资源状态表和所述宿资源状态表有一致的光 突发时隙可用, 则根据所述当前带宽请求的源节点标识和宿节点标识查询路 由表, 如所述路由表中有未选择的路由, 则选取一路由, 根据所选路由依波 长次序查询所述链路资源状态表, 如链路在同一波长和源端宿端有一致的光 突发时隙资源可用,则选取一个满足要求的波长和光突发时隙位置进行分配。
7、 如权利要求 1-6任一项所述方法, 其中,
所述新带宽地图携带每个突发传送通道的波长和光突发时隙配置信息, 所述新带宽地图由一个或多个子地图组成, 每个子地图包含一个波长字段和 一个光突发时隙分配字段。
8、 如权利要求 7所述方法, 其中,
所述波长字段由波长标识和光突发数量组成, 所述光突发时隙分配字段 釆用比特映射方式, 每个比特对应数据帧中的一个光突发时隙位置。
9、 如权利要求 1所述方法, 其中, 所述通过控制帧携带所述新带宽地图 逐跳分发至从节点, 包括:
根据所述新带宽地图进行对象编码, 由控制帧携带编码后的新带宽地图 逐跳分发至从节点。
10、 一种带宽地图更新的装置, 包括初始化模块、 分配模块、 生成模块、 分发模块和更新模块, 其中:
所述初始化模块设置为: 收到控制帧携带的带宽^艮告后, 新建带宽地图, 新建资源状态表, 将新建的资源状态表中所有资源状态置为可用;
所述分配模块设置为: 根据待更新的带宽地图的跨主节点传送通道上路 分配结构, 添加新建的带宽地图跨主节点传送通道下路分配结构, 并触发所 述更新模块;
所述生成模块设置为: 根据控制帧上报的带宽报告, 为当前带宽请求逐 一分配波长和光突发时隙, 加入所述新建的带宽地图, 生成新带宽地图, 并 触发所述更新模块;
所述分发模块, 其设置为: 通过控制帧携带所述新带宽地图逐跳分发至 从节点; 以及
所述更新模块, 其设置为: 受触发后更新所述资源状态表。
11、 如权利要求 10所述装置, 其中,
所述分配模块是设置为以如下方式根据待更新的带宽地图的跨主节点传 送通道上路分配结构,添加新建的带宽地图跨主节点传送通道下路分配结构: 遍历待更新的带宽地图, 将其中跨所述主节点传送通道上路分配结构的波长 和光突发时隙位置, 配置为所述新建的带宽地图中同一传送通道下路分配结 构的波长和光突发时隙位置。
12、 如权利要求 10所述装置, 其中,
所述初始化模块是设置为以如下方式新建资源状态表包括: 新建当前分 配周期的源 /宿资源状态表和当前分配周期的链路资源状态表和下一分配周 期的宿资源状态表, 所述源 /宿资源状态表为 N X I M三维矩阵表, 表示节 点 n第 i个发射机 /接收机在第 m个光突发时隙位置的可用状态, n=l , 2, N; i=l , 2, I; m=l , 2, M; N, I, M均为大于 1 的整数, N表 示 OBTN环网节点数, I表示节点数据通道收发机数量, M表示一个数据帧 的光突发时隙总数; 所述链路资源状态表为 L X W X M三维矩阵表, 表示链 路 1中波长 w在第 m个光突发时隙位置的可用状态, 1=1 , 2, ... ... , L; w=l ,
2, ..· , W; m=l , 2, M; L, W, M均为大于 1的整数, L表示链路数,
W表示数据通道波长数, M表示一个数据帧的光突发时隙总数。
13、 如权利要求 12所述装置, 其中,
所述更新模块是设置为以如下方式更新所述相应的资源状态表: 对于非 跨主节点业务, 更新所述当前分配周期的源 /宿资源状态表和所述当前分配周 期的链路资源状态表; 对于跨主节点上路业务, 更新所述当前分配周期的源 资源状态表、 所述当前分配周期的链路资源状态表和所述下一分配周期的宿 资源状态表; 对于跨主节点下路业务, 更新所述当前分配周期的宿资源状态 表和所述当前分配周期的链路资源状态表。
14、 如权利要求 10所述装置, 其中,
所述生成模块是设置为以如下方式为当前带宽请求逐一分配波长和光突 发时隙: 根据所述当前带宽请求的源节点标识和宿节点标识, 依次查询源资 源状态表和宿资源状态表, 如所述源资源状态表和所述宿资源状态表有一致 的光突发时隙可用, 则根据所述当前带宽请求的源节点标识和宿节点标识查 询路由表, 如所述路由表中有未选择的路由, 则选取一路由, 根据所选路由 依波长次序查询所述链路资源状态表, 如链路在同一波长和源端宿端有一致 的光突发时隙资源可用, 则选取一个满足要求的波长和光突发时隙位置进行 分配。
15、 如权利要求 10-14任一项所述装置, 其中,
所述生成模块 , 生成的新带宽地图携带每个突发传送通道的波长和光突 发时隙配置信息, 所述新带宽地图由一个或多个子地图组成, 每个子地图包 含一个波长字段和一个光突发时隙分配字段, 所述波长字段由波长标识和光 突发数量组成, 所述光突发时隙分配字段釆用比特映射方式, 每个比特对应 数据帧中的一个光突发时隙位置。
16、 如权利要求 10所述装置, 其中,
所述分发模块是设置为根据所述新带宽地图进行对象编码, 由控制帧携 带编码后的新带宽地图逐跳分发至从节点。
17、 一种计算机程序, 包括程序指令, 当该程序指令被主节点执行时, 使得该主节点可执行权利要求 1-9任一项所述的方法。
18、 一种载有权利要求 17所述计算机程序的载体。
PCT/CN2014/079966 2013-08-20 2014-06-16 一种带宽地图更新的方法及装置 WO2014180447A1 (zh)

Priority Applications (3)

Application Number Priority Date Filing Date Title
EP14794041.5A EP3038279B1 (en) 2013-08-20 2014-06-16 Bandwidth map update method and device
JP2016535314A JP6373996B2 (ja) 2013-08-20 2014-06-16 帯域幅マップ更新方法及び装置
US14/912,930 US9698930B2 (en) 2013-08-20 2014-06-16 Bandwidth map update method and device

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201310365467.1A CN104427412B (zh) 2013-08-20 2013-08-20 一种带宽地图更新的方法及装置
CN201310365467.1 2013-08-20

Publications (1)

Publication Number Publication Date
WO2014180447A1 true WO2014180447A1 (zh) 2014-11-13

Family

ID=51866827

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2014/079966 WO2014180447A1 (zh) 2013-08-20 2014-06-16 一种带宽地图更新的方法及装置

Country Status (5)

Country Link
US (1) US9698930B2 (zh)
EP (1) EP3038279B1 (zh)
JP (1) JP6373996B2 (zh)
CN (1) CN104427412B (zh)
WO (1) WO2014180447A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113992259A (zh) * 2021-10-22 2022-01-28 中国人民解放军63921部队 一种时隙资源拓展图的构建方法

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105744385B (zh) * 2014-12-10 2020-06-16 中兴通讯股份有限公司 一种光突发传送网的传输方法和系统
JP7035877B2 (ja) * 2018-07-20 2022-03-15 富士通株式会社 伝送装置
JP7307388B2 (ja) * 2020-02-28 2023-07-12 日本電信電話株式会社 光コア部を有する光ネットワーク、タイムスロット化した受信を使用するノード
EP4138317A1 (en) * 2021-08-19 2023-02-22 Nokia Solutions and Networks Oy Adaptive channel loading in wdm optical communication systems

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101656894A (zh) * 2008-08-20 2010-02-24 华为技术有限公司 包分插复用设备及包分插复用设备的数据传输方法
CN101895367A (zh) * 2009-05-22 2010-11-24 华为技术有限公司 一种数据传输方法、网络设备和系统
CN102025478A (zh) * 2009-09-15 2011-04-20 华为技术有限公司 数据传送、接收的方法及装置

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB9817270D0 (en) * 1998-08-07 1998-10-07 Northern Telecom Ltd A method of allocating resources in a telecommunications network
JP3516972B2 (ja) * 1993-04-22 2004-04-05 株式会社東芝 通信システム
CA2339902A1 (en) * 2001-03-07 2002-09-07 Cedric Don-Carolis Photonic communication system with sub-"line rate" bandwidth granularity
US7009991B2 (en) * 2002-03-28 2006-03-07 Matisse Networks Reservation-based media access controller and reservation-based optical network
US7680041B2 (en) * 2006-01-31 2010-03-16 Zensys A/S Node repair in a mesh network
CN101753415B (zh) * 2008-12-08 2012-01-04 华为技术有限公司 带宽地图更新的方法和装置
CN101753527B (zh) * 2008-12-22 2012-10-17 华为技术有限公司 传送网中传送带宽分配信息的方法、系统及设备
CN102006232B (zh) * 2010-11-26 2012-07-04 华为技术有限公司 一种动态带宽分配方法、装置和系统

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101656894A (zh) * 2008-08-20 2010-02-24 华为技术有限公司 包分插复用设备及包分插复用设备的数据传输方法
CN101895367A (zh) * 2009-05-22 2010-11-24 华为技术有限公司 一种数据传输方法、网络设备和系统
CN102025478A (zh) * 2009-09-15 2011-04-20 华为技术有限公司 数据传送、接收的方法及装置

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3038279A4 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113992259A (zh) * 2021-10-22 2022-01-28 中国人民解放军63921部队 一种时隙资源拓展图的构建方法
CN113992259B (zh) * 2021-10-22 2023-05-16 中国人民解放军63921部队 一种时隙资源拓展图的构建方法

Also Published As

Publication number Publication date
CN104427412A (zh) 2015-03-18
JP6373996B2 (ja) 2018-08-15
EP3038279A1 (en) 2016-06-29
JP2016531510A (ja) 2016-10-06
EP3038279B1 (en) 2017-08-02
CN104427412B (zh) 2019-06-18
US9698930B2 (en) 2017-07-04
EP3038279A4 (en) 2016-10-05
US20160197699A1 (en) 2016-07-07

Similar Documents

Publication Publication Date Title
CN1285241C (zh) 波分复用光突发交换网络中的数据与控制调度方法和装置
US7272310B2 (en) Generic multi-protocol label switching (GMPLS)-based label space architecture for optical switched networks
US20120207467A1 (en) Path Computation Element System and Method of Routing and Wavelength Assignment in a Wavelength Switched Optical Network
EP3443693B1 (en) Channel bonding in multiple-wavelength passive optical networks (pons)
WO2014180447A1 (zh) 一种带宽地图更新的方法及装置
CN107222404B (zh) Woban中带有冗余编码的并行路径转发方法
CN103581006A (zh) 用于灵活栅格光网络全局优化的系统架构及其全局优化方法
JP2014090408A (ja) パケット光ネットワークの宛先ノードに外部の光データパケットを提供する方法及びシステム
CN105959163A (zh) 一种基于软件定义的无源光互连网络结构及数据通信方法
Chunming et al. Polymorphic Control for Cost‐Effective Design of Optical Networks
CN100548067C (zh) 一种在多粒度光交换网络中实现光通道建立的控制方法
US10206018B2 (en) Transmission method and system for optical burst transport network
WO2014187425A1 (zh) 跨主节点业务处理方法和装置
Martinez et al. Experimental GMPLS routing for dynamic provisioning in translucent wavelength switched optical networks
Jin et al. Study of dynamic routing and spectrum assignment schemes in bandwidth flexible optical networks
CN108737911B (zh) 基于广义光突发交换的potn业务承载方法及系统
JP6753793B2 (ja) 帯域割当装置
WO2016101827A1 (zh) 一种光突发传送网业务跨环的实现方法和相应的主节点
CN104468409A (zh) 一种基于非固定多线程轮询的动态带宽分配方法
Radzi et al. Dynamic bandwidth allocation EPON survey
WO2019080744A1 (zh) 控制通道实现方法、装置、设备、存储介质和处理方法
Guo et al. A Novel Proportional Window based Dyamic Bandwidth Allocation Algorithm in Optical Network
Kim et al. Property of each area technique for network survivability in optical networks
KR100898344B1 (ko) 광전송 망에서 서브 도메인 가중치를 이용한 분산 자원공유 방법
CN104170292B (zh) 光同轴网络架构中传输数据、资源调度的方法和装置

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: 14794041

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2016535314

Country of ref document: JP

Kind code of ref document: A

WWE Wipo information: entry into national phase

Ref document number: 14912930

Country of ref document: US

NENP Non-entry into the national phase

Ref country code: DE

REEP Request for entry into the european phase

Ref document number: 2014794041

Country of ref document: EP

WWE Wipo information: entry into national phase

Ref document number: 2014794041

Country of ref document: EP