EP2497213A1 - Method and device for processing data in a network element - Google Patents

Method and device for processing data in a network element

Info

Publication number
EP2497213A1
EP2497213A1 EP09749074A EP09749074A EP2497213A1 EP 2497213 A1 EP2497213 A1 EP 2497213A1 EP 09749074 A EP09749074 A EP 09749074A EP 09749074 A EP09749074 A EP 09749074A EP 2497213 A1 EP2497213 A1 EP 2497213A1
Authority
EP
European Patent Office
Prior art keywords
slots
data stream
node
network element
network
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.)
Withdrawn
Application number
EP09749074A
Other languages
German (de)
French (fr)
Inventor
Claus Gruber
Dominic Schupke
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.)
Xieon Networks SARL
Original Assignee
Nokia Siemens Networks Oy
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 Nokia Siemens Networks Oy filed Critical Nokia Siemens Networks Oy
Publication of EP2497213A1 publication Critical patent/EP2497213A1/en
Withdrawn legal-status Critical Current

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04J—MULTIPLEX COMMUNICATION
    • H04J3/00—Time-division multiplex systems
    • H04J3/16—Time-division multiplex systems in which the time allocation to individual channels within a transmission cycle is variable, e.g. to accommodate varying complexity of signals, to vary number of channels transmitted
    • H04J3/1682—Allocation of channels according to the instantaneous demands of the users, e.g. concentrated multiplexers, statistical multiplexers

Definitions

  • the invention relates to a method and to a device for proces ⁇ sing data in a network element and to a communication system comprising at least one such device.
  • This invention in particular relates to the field of communi- cation networks, e.g., packet and/or TDM switching and rou ⁇ ting approaches in a network.
  • communi- cation networks e.g., packet and/or TDM switching and rou ⁇ ting approaches in a network.
  • IP or Ethernet packets Today, almost all data generated by applications is packet- based data comprising IP or Ethernet packets, wherein packets vary in size.
  • packets are sent asynchronously between routers and various bitrates apply to end-to-end communication.
  • the network comprises several network nodes, which may buffer and schedu ⁇ le packets in order to avoid or reduce congestion and/or col- lisions.
  • the problem to be solved is to overcome the disadvantages in- dicated above and in particular to provide flexible or ar ⁇ bitrary bitrates in an end-to-end relation.
  • the network element provides at least one
  • outgoing data stream wherein said outgoing data stream comprises transport slots
  • transport slots of the at least one out ⁇ going data stream are configured according to a flexible bitrate.
  • bitrate can be changed or modified during operation, e.g., it might be switched from 1.5 Gbps to 1 Gbps or vice versa.
  • Said data stream may comprise data packets of at least one communication path, in particular several communication paths can be time-multiplexed via one data stream. It is also pos ⁇ sible that one communication path utilizes several slots of the data stream; in this case the different slot assignments multiplexed via the data stream logically correspond to dif ⁇ ferent "channels" (which may be referred to as virtual chan ⁇ nels) . In other words, a data stream may convey several mul ⁇ tiplexed channels.
  • the network element may comprise ingoing and outgoing interfaces, which may be connected to another network element each. Traffic can be conveyed via said connections and arrive at the ingoing interfaces of the network element.
  • the network element processes the traffic that arrives at its ingoing in ⁇ terfaces and forwards it via its outgoing interfaces.
  • the traffic may be routed according to the configuration provided to or at the network element. As an option, a portion of the ingoing traffic may also terminate at the network element in case this network element is the destination of such ingoing traffic .
  • said transport slots are TDM transport slots .
  • transport slots are ap ⁇ plicable as well, e.g., frequency-multiplexed, code- multiplexed transport slots or combinations of the ones men ⁇ tioned herein.
  • the transport slots of the at least one outgoing data stream are configured according to a fle ⁇ xible bitrate at least partially based on at least one of the following criteria:
  • the transport slots of the at least one outgoing data stream are configured based on a preconfi- gured, anticipated or predicted time-slot assignment.
  • the time-slot assignment can be set, e.g., by a net- work management system or it can be communicated to the net ⁇ work element on various communication paths (in-band signa ⁇ ling or out-of-band signaling) .
  • the network element can be re-adjusted if need would be.
  • the network element may be able to predict or anticipate the time-slot assignment from a past utilization of time slots for a particular data stream.
  • the time-slot assignment is conducted in the network element prior to sending any data packets.
  • the at least one outgoing data stream is configured via a network management system.
  • transport slots of the at least one outgoing data stream are configured based on a reactive time-slot assignment.
  • bitrates of and along end-to-end data streams (paths) and the required time- slot lengths might not be known in advance.
  • the information required may be conveyed together with the inco- ming data.
  • At least one incoming data packet comprises information regarding a time slot assignment to be utilized by the network element.
  • transport slots of the at least one outgoing data stream are configured according to a fle ⁇ xible bitrate based on an order of incoming data packets and an assignment of the incoming data packets to outgoing data streams, wherein such assignment is conveyed with the data packets .
  • the network element is a no ⁇ de of a network, in particular a node of at least one commu ⁇ nication path.
  • the network element conveys data of a communication via several transport slots of at least two logical channels.
  • the network element may utilize different logical channels of the multiplexed data stream to convey data of a single communication.
  • the time slots for the different channels may vary in size. This allows a high de ⁇ gree of flexibility to convey the communication (data, speech, IP-traffic, or the like) between an origin and its destination .
  • the network element conveys data of a communication via separate outgoing data streams.
  • different routes can be utilized to convey the data of a single communication:
  • the data packets are assigned to different transport slots that are conveyed via different paths through the network.
  • a device compri- sing or being associated with a processing unit that is arranged such that the method as described herein is executable thereon .
  • Said processing unit may comprise at least one of the follo- wing: a processor, a microcontroller, a hard-wired circuit, an ASIC, an FPGA, a logic device.
  • the device is a network ele ⁇ ment, in particular a node of a communication network, wherein said node is in particular a portion of at least one communication path.
  • the device may process incoming and outgoing packet-oriented data, wherein outgoing data may be multiplexed in various transport slots utilizing different paths throughout the net ⁇ work.
  • the device by processing the incoming data may utilize the outgoing data stream pursuant to a flexible bitrate as also indicated above.
  • Embodiments of the invention are shown and illustrated in the following figures: shows a schematic diagram of several nodes of a net ⁇ work providing a flexible TDM concept; shows a schematic diagram comprising a portion of a network with one network node, wherein data packets arrive via a first path and via a second path at this node and said node combines such packets into an out ⁇ going data stream; shows a schematic diagram of several nodes of a net ⁇ work based on Fig.l utilizing several time slots per time unit of a first communication to be conveyed, as well as utilizing two sub-paths for conveying a sec ⁇ ond communication;
  • Fig.4 shows a schematic diagram comprising a portion of a network with a first node comprising three interfaces and a second node with one interface visualizing a reactive time slot handling with and without path in ⁇ formation included
  • Fig.5 shows a schematic block diagram of a network compris ⁇ ing a single node with three ports as well as a switching table
  • Fig.6 shows a block diagram comprising an outer slot with a header and several inner slots, each comprising a header of its own.
  • the actual bandwidth bet ⁇ ween two end-points is often known in advance (e.g., 10Mbps via a leased line) .
  • Several data streams (paths) have to be conveyed between different edge devices of the network.
  • the delivery of data of all paths may be guaranteed with a cer ⁇ tain quality of service (e.g., bandwidth, delay, jitter, etc . ) .
  • the approach presented herein in particular provides a pre- configured TDM assignment at a preferable (e.g., minimal) granularity of bit level, to offer transport of any client data size without additional transport framing packet
  • TDM transport slots can be assigned to dif ⁇ ferent paths in network nodes and/or to their interfaces.
  • an arbitrary time-slot duration and (based on such time-slot duration) an arbitrary bitrate can be configured (this is also referred to herein as flexible TDM) .
  • the avai ⁇ lable bitrate on an interface and/or link can be divided into several slots in order to allow conveying different data streams .
  • time- slots and time-slot assignments can be determined and confi ⁇ gured in the nodes of the path before sending the packets.
  • Fig.l shows a schematic diagram of several nodes of a network visualizing a flexible TDM concept.
  • the network comprises no ⁇ des A to G.
  • Three paths I to III are to be established. Path I leads from node A via nodes C and E to node G, path II leads from node B via nodes C and E to node F and path III leads from node D via node E to node G.
  • the bitrates for paths I to III are: 30 Gbps, 70 Gbps and 60 Gbps .
  • the nodes A to G are configured by a network management sys ⁇ tem NMS .
  • all interfaces are capable of processing a bitrate of 100 Gbps.
  • the available bitrate on an interface and/or link is divided into several slots in order to allow sending different data streams.
  • a time unit e.g., a second
  • 3/10 of the time unit are used for transporting data of path I and 7/10 of the time unit are used for transporting data of path II.
  • node E conveys the first time-slot (3/10 of the time unit) towards node G and the second time-slot (7/10 of the time-unit) towards node F.
  • the assignment of time-slots can be synchroni ⁇ zed between all network nodes so that no buffering of data is needed in intermediate nodes.
  • the buffer-duration may remain fixed .
  • Fig.2 shows a schematic diagram comprising a portion of a network with one network node 201.
  • Data packets 202 arrive via a first path 203 and data packets 204 arrive via a second path 205 at the node 201.
  • a data stream 206 combines both pa- ckets 202 and 204 and is conveyed from the node 201 via a path 207.
  • the transmission delay and the required buffer size of the node 201 can be determined.
  • the packets 202 are buffered to correspond to packets 208 that are then packed into the data stream 206.
  • end-to-end quality of service still can be guaranteed.
  • the transmission can be performed continuously using one slot or it can be distributed onto several time-slots as shown in Fig.3 for path I (equal time-slots) and path III (different time-slots) .
  • Fig.3 is based on the network according to Fig.l. However, the path I utilizes se- veral time slots per time unit of the same traffic to be con ⁇ veyed, i.e.
  • Path II comprises two sub-paths 303 and 304 conveying data from node B to node F, wherein the sub-path 303 corresponds to the path II of Fig.l and the sub-path 304 leads via nodes C, D and E to node F.
  • the sub-path 303 con ⁇ veys data at a bitrate of 30 Gbps and the sub-path 304 con ⁇ veys data at a bitrate of 40 Gbps.
  • Traffic from different time-slots is reordered at the edge nodes or at nodes where paths are joining, in particular at nodes C, D and E.
  • Multipath transmission and reordering can be done similarly to VCAT mechanisms as known from ITU-T G.707.
  • An assignment of time-slots to single or multiple paths as well as the configuration of the network nodes can be done from a centralized entity, e.g., the NMS (shown in Fig.l and Fig.3, which can be partly or fully distributed (e.g., using control plane messages) .
  • time-slot assignment could be an objective of the time-slot assignment to pro ⁇ vide a fair distribution of bitrates or to guarantee re- quested bitrates, to provide a minimum delay inside a node, a minimum end-to-end delay, minimum transport costs, or the li ⁇ ke. Also, combinations of several objectives could be met.
  • time-slot can be static for a given period of time or it can be changed frequently to adapt to known or anticipated changes of bitrate.
  • Reactive time-slot assignment In the reactive time-slot assignment mode, the bitrates of and along end-to-end data streams (paths) and the required time-slot lengths are not known in advance.
  • Fig.4 shows a schematic diagram comprising a portion of a network with a first node 401 comprising three interfaces 402, 403 and 404 and a second node 405 with an interface 406.
  • a path 1 leads via the interface 402 to the interface 403 and a path 2 leads from the interface 402 to the interface 404 and thus towards the interface 406 of the node 405.
  • the interface 402 is configured as shown in a table 407 of Fig.4:
  • the first time slot is associated with path 1
  • the se ⁇ cond time slot is associated with path 2
  • the third time slot is again associated with path 1 and so on.
  • the path 1 is associated with interface 403
  • the path 2 is associated with interface 404 as outgoing interface.
  • Fig.4 also shows how an incoming stream of data arriving at said interface 402 is distributed among interfaces 403 and 404.
  • the table 407 may be omitted in case a data pa ⁇ cket itself indicates the path. This is shown for a data pa- cket 408 that is conveyed to the interface 406, said data pa ⁇ cket 408 comprising not only a duration 409 (indicating the length of the data packet) , but also a path information 410 telling the node 405 towards which path (not shown) the inco ⁇ ming packet 408 needs to be conveyed.
  • time-slot (frame/container) can be detected via (deep- ) packet inspection. Hence, no information about the length of the time-slot needs to be included in the frame or container. However, additional detection mechanisms are required in such scenario.
  • the path information may also be provided in the frame or container.
  • the switching to the appropriate outgoing interface can be performed in the reactive mode.
  • a header of a flexible TDM slot may comprise at least one of the following:
  • Time-synchronization information Information needed to synchronize the clocks of each node (e.g., similar to IEEE 1588) ;
  • the time slots can carry further overhead information, e.g., tracing overhead that can be used in nodes to detect misconfigurations of slot assignments; the tracing overhead can be updated by intermediate nodes;
  • Additional flags can be included to, e.g., enable
  • slots of one path have equal length
  • the in ⁇ formation about paths and slot-lengths can be preconfigured in each intermediate node. Since each node knows at each time which slot comes next, what length the slot has and where to switch to, no additional header information is required. Slot lengths and path sequence:
  • slot-length and path sequence can change over time or with each cycle.
  • the change can be configu- red in advance.
  • Fig.5 shows a portion of a network comprising a single node 501 with three ports I, II and III as well as a switching table 502 and a switching table 503.
  • Table 502 shows an exemplary switching for the port I with fixed slot lengths and sequences
  • table 503 shows an exemplary switching for the port I with a change of slot lengths and changing se ⁇ quences.
  • the duration of the slots can be adapted .
  • a duration of the flexible slots, a path sequence and a route can be configured via a centralized or via a decentralized approach.
  • the centralized approach can be performed, e.g., by a Network Management System (NMS) .
  • NMS Network Management System
  • the NMS may be aware of traffic requests as well as the network topology. After calculating the routes for each traffic (or portion thereof) , the NMS assigns slot durations and sequences for each path.
  • Each node can be configured by the NMS via ap ⁇ basementte protocols, e.g., by using SNMP .
  • the decentralized approach can be performed similar to an ap ⁇ proach configuring MPLS paths with RSVP-TE: When a node decides to establish or to change a path, a reservation message can be sent from the source node to the destination node.
  • the path can be routed at the source or information may be sent hop-by-hop to the destination node.
  • the reservation request message can include information about the path characte ⁇ ristics, e.g., required bandwidth, maximum delay, etc.
  • a confirmation message is sent in backward directi ⁇ on.
  • the confirmation message triggers the actual reservation of bandwidth.
  • slot lengths and preferred slot times can be proposed in the response message. Upstream nodes can configure their switching behavior accordingly. It is noted that the slot lengths and path sequences may have local link characteristics or can be identical for the whole path.
  • slots can be sub-divided to carry traffic of smaller slots.
  • This mechanism can be used to speed up forwarding of traffic if many small traffic portions are transpor ⁇ ted via common sub-paths.
  • an encapsulation and de- capsulation can be conducted at the outer path end-nodes.
  • ⁇ formation about the sub-slot structure can be included in the outer slot header or can be preconfigured in the end-nodes.
  • Fig.6 shows an outer slot 601 with a header and several inner slots 602 to 606, each comprising a header of its own.
  • the approach suggested allows for a higher degree of flexibility as any granularity of bitrates can be selec- ted, e.g., 893827 bps or 1 Gbps . This further allows for a less complex operation without any need for mapping of client data to a fixed container size. Also, as the overhead is significantly reduced, this approach is bandwidth efficient. ODU traffic can be transported transparently by the approach provided. In particular, the approach suggested may utilize an optical transport layer for conveying traffic.
  • nodes of the network may pro ⁇ vide services of ODU clients. This is in particular useful for an end-to-end relation in case both end nodes support ODU traffic and flexible TDM traffic as described herein.
  • a remapping of ODU traffic into any slot-size of the flexible TDM suggested can be done at intermediate nodes if required (e.g. for using multipath) .
  • the same ODU format may be reestablished at the other side of the flexible TDM domain.

Landscapes

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

Abstract

A method and a device for processing data in a network element are provided, wherein the network element provides at least one outgoing data stream, wherein said outgoing data stream comprises transport slots, and wherein the transport slots of the at least one outgoing data stream are configured according to a flexible bitrate. In addition, a communication system comprising at least one such device is suggested.

Description

Description
Method and device for processing data in a network element The invention relates to a method and to a device for proces¬ sing data in a network element and to a communication system comprising at least one such device.
This invention in particular relates to the field of communi- cation networks, e.g., packet and/or TDM switching and rou¬ ting approaches in a network.
Today, almost all data generated by applications is packet- based data comprising IP or Ethernet packets, wherein packets vary in size. In packet-routed and packet-switched networks, packets are sent asynchronously between routers and various bitrates apply to end-to-end communication. The network comprises several network nodes, which may buffer and schedu¬ le packets in order to avoid or reduce congestion and/or col- lisions.
Present approaches comprise complex and hence costly schedu¬ ling mechanisms requiring powerful and hence expensive hard¬ ware to ensure the required throughput, to reduce delay vari- ations and to achieve a decent quality of service. However, such quality of service can often not be guaranteed for indi¬ vidual end-to-end relations, in particular with regard to high load situations of high priority traffic. In TDM networks like SDH/Sonet and OTN/OTH, throughput can be guaranteed for individual end-to-end relations and delay va¬ riations are negligible. However, these networks need to have an embedded static framing mechanism and framing hierarchies (e.g., VCs in STM) . These frames are based on constant bitra- tes and are difficult to be mapped from lower layer frames to higher layer frames. Dependent on the mapping into frames, a significant overhead cannot be avoided (e.g., transmitting a client data rate of 70 Mbps as 2 x VC3s with a filling ratio of only 73% according to ITU-T G.707).
The problem to be solved is to overcome the disadvantages in- dicated above and in particular to provide flexible or ar¬ bitrary bitrates in an end-to-end relation.
This problem is solved according to the features of the inde¬ pendent claims. Further embodiments result from the depending claims.
In order to overcome this problem, a method is suggested for processing data in a network element,
- wherein the network element provides at least one
outgoing data stream, wherein said outgoing data stream comprises transport slots,
- wherein the transport slots of the at least one out¬ going data stream are configured according to a flexible bitrate.
This approach allows utilizing a fine granularity of bitra¬ tes, in particular to adjust a bitrate of an outgoing data stream. Hence, the slot size and thus bitrate is no longer fixed for a particular data stream (or a combination of data streams) , but can be adjusted such that packets of this par¬ ticular data stream fit to the slots reserved. Hence, this approach significantly reduces any overhead and waste of bandwidth and allows for an efficient data transmission. The bitrate can be changed or modified during operation, e.g., it might be switched from 1.5 Gbps to 1 Gbps or vice versa.
With the transport slots being configured, there may be no need for deep-packet inspection. Switching of ingoing traffic to said transport slots of the outgoing data stream can be provided in an efficient manner without having to look into the packets of the ingoing traffic. This leads to less pro¬ cessing complexity at the network element. Said data stream may comprise data packets of at least one communication path, in particular several communication paths can be time-multiplexed via one data stream. It is also pos¬ sible that one communication path utilizes several slots of the data stream; in this case the different slot assignments multiplexed via the data stream logically correspond to dif¬ ferent "channels" (which may be referred to as virtual chan¬ nels) . In other words, a data stream may convey several mul¬ tiplexed channels.
The network element may comprise ingoing and outgoing interfaces, which may be connected to another network element each. Traffic can be conveyed via said connections and arrive at the ingoing interfaces of the network element. The network element processes the traffic that arrives at its ingoing in¬ terfaces and forwards it via its outgoing interfaces. The traffic may be routed according to the configuration provided to or at the network element. As an option, a portion of the ingoing traffic may also terminate at the network element in case this network element is the destination of such ingoing traffic .
In an embodiment, said transport slots are TDM transport slots .
It is noted that also other sorts of transport slots are ap¬ plicable as well, e.g., frequency-multiplexed, code- multiplexed transport slots or combinations of the ones men¬ tioned herein.
In another embodiment, the transport slots of the at least one outgoing data stream are configured according to a fle¬ xible bitrate at least partially based on at least one of the following criteria:
- to guarantee a requested bitrate for a particular
service ;
- to provide a reduced end-to-end delay;
- to ensure reduced transport costs. Hence, the criteria mentioned may be utilized or optimized by adjusting the transport slots accordingly. In a further embodiment, the transport slots of the at least one outgoing data stream are configured based on a preconfi- gured, anticipated or predicted time-slot assignment.
Hence, the time-slot assignment can be set, e.g., by a net- work management system or it can be communicated to the net¬ work element on various communication paths (in-band signa¬ ling or out-of-band signaling) . The network element can be re-adjusted if need would be. Also, the network element may be able to predict or anticipate the time-slot assignment from a past utilization of time slots for a particular data stream. Advantageously, the time-slot assignment is conducted in the network element prior to sending any data packets.
In a next embodiment, the at least one outgoing data stream is configured via a network management system.
It is also an embodiment that the transport slots of the at least one outgoing data stream are configured based on a reactive time-slot assignment.
In the reactive time-slot assignment mode, bitrates of and along end-to-end data streams (paths) and the required time- slot lengths might not be known in advance. In this case, the information required may be conveyed together with the inco- ming data.
Pursuant to another embodiment, at least one incoming data packet comprises information regarding a time slot assignment to be utilized by the network element.
According to an embodiment, transport slots of the at least one outgoing data stream are configured according to a fle¬ xible bitrate based on an order of incoming data packets and an assignment of the incoming data packets to outgoing data streams, wherein such assignment is conveyed with the data packets . According to another embodiment, the network element is a no¬ de of a network, in particular a node of at least one commu¬ nication path.
In yet another embodiment, the network element conveys data of a communication via several transport slots of at least two logical channels.
Hence, the network element may utilize different logical channels of the multiplexed data stream to convey data of a single communication. In particular, the time slots for the different channels may vary in size. This allows a high de¬ gree of flexibility to convey the communication (data, speech, IP-traffic, or the like) between an origin and its destination .
According to a next embodiment, the network element conveys data of a communication via separate outgoing data streams.
In this case, different routes can be utilized to convey the data of a single communication: The data packets are assigned to different transport slots that are conveyed via different paths through the network.
The problem stated above is also solved by a device compri- sing or being associated with a processing unit that is arranged such that the method as described herein is executable thereon .
Said processing unit may comprise at least one of the follo- wing: a processor, a microcontroller, a hard-wired circuit, an ASIC, an FPGA, a logic device. Pursuant to yet an embodiment, the device is a network ele¬ ment, in particular a node of a communication network, wherein said node is in particular a portion of at least one communication path.
The device may process incoming and outgoing packet-oriented data, wherein outgoing data may be multiplexed in various transport slots utilizing different paths throughout the net¬ work. The device by processing the incoming data may utilize the outgoing data stream pursuant to a flexible bitrate as also indicated above.
The problem stated supra is further solved by a communication system comprising at least one device as described herein.
Embodiments of the invention are shown and illustrated in the following figures: shows a schematic diagram of several nodes of a net¬ work providing a flexible TDM concept; shows a schematic diagram comprising a portion of a network with one network node, wherein data packets arrive via a first path and via a second path at this node and said node combines such packets into an out¬ going data stream; shows a schematic diagram of several nodes of a net¬ work based on Fig.l utilizing several time slots per time unit of a first communication to be conveyed, as well as utilizing two sub-paths for conveying a sec¬ ond communication;
Fig.4 shows a schematic diagram comprising a portion of a network with a first node comprising three interfaces and a second node with one interface visualizing a reactive time slot handling with and without path in¬ formation included; Fig.5 shows a schematic block diagram of a network compris¬ ing a single node with three ports as well as a switching table;
Fig.6 shows a block diagram comprising an outer slot with a header and several inner slots, each comprising a header of its own.
In existing (transport) networks, the actual bandwidth bet¬ ween two end-points is often known in advance (e.g., 10Mbps via a leased line) . Several data streams (paths) have to be conveyed between different edge devices of the network. The delivery of data of all paths may be guaranteed with a cer¬ tain quality of service (e.g., bandwidth, delay, jitter, etc . ) .
The approach presented herein in particular provides a pre- configured TDM assignment at a preferable (e.g., minimal) granularity of bit level, to offer transport of any client data size without additional transport framing packet
overhead. Hence, TDM transport slots can be assigned to dif¬ ferent paths in network nodes and/or to their interfaces. In contrast to fixed time-slots as known from conventional TDM networks, an arbitrary time-slot duration and (based on such time-slot duration) an arbitrary bitrate can be configured (this is also referred to herein as flexible TDM) . The avai¬ lable bitrate on an interface and/or link can be divided into several slots in order to allow conveying different data streams .
For example, the following modes are suggested and will be described in more detail hereinafter:
- a preconfigured time-slot assignment;
- a reactive time-slot assignment. Preconfigured time-slot assignment:
In the preconfigured time-slot assignment mode, the actual bitrates of the end-to-end data streams (paths) are known and/or anticipated in advance. With this information, time- slots and time-slot assignments can be determined and confi¬ gured in the nodes of the path before sending the packets.
Fig.l shows a schematic diagram of several nodes of a network visualizing a flexible TDM concept. The network comprises no¬ des A to G. Three paths I to III are to be established. Path I leads from node A via nodes C and E to node G, path II leads from node B via nodes C and E to node F and path III leads from node D via node E to node G. The bitrates for paths I to III are: 30 Gbps, 70 Gbps and 60 Gbps .
The nodes A to G are configured by a network management sys¬ tem NMS . For reasons of illustration, all interfaces are capable of processing a bitrate of 100 Gbps. The available bitrate on an interface and/or link is divided into several slots in order to allow sending different data streams. For example, on a link between nodes C and E a time unit (e.g., a second) is divided into two parts and 3/10 of the time unit are used for transporting data of path I and 7/10 of the time unit are used for transporting data of path II.
As scheduling of the time-slots is known in advance, a node of a network can switch the traffic accordingly without any necessity for deep packet inspection as it would otherwise be necessary in packet switched networks. According to the example, node E conveys the first time-slot (3/10 of the time unit) towards node G and the second time-slot (7/10 of the time-unit) towards node F.
As an option, the assignment of time-slots can be synchroni¬ zed between all network nodes so that no buffering of data is needed in intermediate nodes. However, even if it is not pos¬ sible to transport data without buffering in at least one network node of the path, the buffer-duration may remain fixed .
Fig.2 shows a schematic diagram comprising a portion of a network with one network node 201. Data packets 202 arrive via a first path 203 and data packets 204 arrive via a second path 205 at the node 201. A data stream 206 combines both pa- ckets 202 and 204 and is conveyed from the node 201 via a path 207.
Based on the slots available in the data stream 206, the transmission delay and the required buffer size of the node 201 can be determined. Hence, the packets 202 are buffered to correspond to packets 208 that are then packed into the data stream 206. Advantageously, end-to-end quality of service still can be guaranteed. As another option, the transmission can be performed continuously using one slot or it can be distributed onto several time-slots as shown in Fig.3 for path I (equal time-slots) and path III (different time-slots) . Fig.3 is based on the network according to Fig.l. However, the path I utilizes se- veral time slots per time unit of the same traffic to be con¬ veyed, i.e. data packets 301 and 302 are conveyed via said path I. Path II, however, comprises two sub-paths 303 and 304 conveying data from node B to node F, wherein the sub-path 303 corresponds to the path II of Fig.l and the sub-path 304 leads via nodes C, D and E to node F. The sub-path 303 con¬ veys data at a bitrate of 30 Gbps and the sub-path 304 con¬ veys data at a bitrate of 40 Gbps.
Traffic from different time-slots is reordered at the edge nodes or at nodes where paths are joining, in particular at nodes C, D and E. Multipath transmission and reordering can be done similarly to VCAT mechanisms as known from ITU-T G.707. An assignment of time-slots to single or multiple paths as well as the configuration of the network nodes can be done from a centralized entity, e.g., the NMS (shown in Fig.l and Fig.3, which can be partly or fully distributed (e.g., using control plane messages) .
It could be an objective of the time-slot assignment to pro¬ vide a fair distribution of bitrates or to guarantee re- quested bitrates, to provide a minimum delay inside a node, a minimum end-to-end delay, minimum transport costs, or the li¬ ke. Also, combinations of several objectives could be met.
It is another option that the time-slot can be static for a given period of time or it can be changed frequently to adapt to known or anticipated changes of bitrate.
Reactive time-slot assignment: In the reactive time-slot assignment mode, the bitrates of and along end-to-end data streams (paths) and the required time-slot lengths are not known in advance.
However, the number of paths, the order of time-slots and their assignments to paths may be known in advance. Hence, the intermediate nodes know that a next time-slot is associa¬ ted with a certain path. At the outgoing interface of the intermediate node, the length of the time-slot is not known be¬ forehand. The size of the time slot can be provided with a header of the frame or container. Thus, the appropriate swit¬ ching to outgoing interfaces can be done according to the pre-determined path information and the duration determined by the header information. Fig.4 shows a schematic diagram comprising a portion of a network with a first node 401 comprising three interfaces 402, 403 and 404 and a second node 405 with an interface 406. A path 1 leads via the interface 402 to the interface 403 and a path 2 leads from the interface 402 to the interface 404 and thus towards the interface 406 of the node 405.
The interface 402 is configured as shown in a table 407 of Fig.4: The first time slot is associated with path 1, the se¬ cond time slot is associated with path 2 and the third time slot is again associated with path 1 and so on. The path 1 is associated with interface 403 and the path 2 is associated with interface 404 as outgoing interface. Fig.4 also shows how an incoming stream of data arriving at said interface 402 is distributed among interfaces 403 and 404.
As an option, the table 407 may be omitted in case a data pa¬ cket itself indicates the path. This is shown for a data pa- cket 408 that is conveyed to the interface 406, said data pa¬ cket 408 comprising not only a duration 409 (indicating the length of the data packet) , but also a path information 410 telling the node 405 towards which path (not shown) the inco¬ ming packet 408 needs to be conveyed.
The beginning of a time-slot (frame/container) can be detected via (deep- ) packet inspection. Hence, no information about the length of the time-slot needs to be included in the frame or container. However, additional detection mechanisms are required in such scenario.
The path information may also be provided in the frame or container. Thus, in addition to the duration also the switching to the appropriate outgoing interface can be performed in the reactive mode.
Instead of paths the assignment of a time-slot sequence to an outgoing interface may be determined or known. It is also an option that instead of paths the assignment of a time-slot sequence to a destination is known. Header Information:
A header of a flexible TDM slot may comprise at least one of the following:
- Length (in bytes) or duration (in ms or less) of the slot ;
- Destination address or identification of a path;
- Source address of a path;
- Path identifier and/or label: A unique identifier
that identifies the path the slot belongs to;
- Slot number that is incremented for each slot used or sent ;
- Additional information regarding quality of service or availability characteristics, e.g., an end-to-end delay required;
- Time-synchronization information: Information needed to synchronize the clocks of each node (e.g., similar to IEEE 1588) ;
- Tracing information: The time slots can carry further overhead information, e.g., tracing overhead that can be used in nodes to detect misconfigurations of slot assignments; the tracing overhead can be updated by intermediate nodes;
- Additional flags can be included to, e.g., enable
tracing;
- Information about a sub-slot structure.
It is noted that according to an exemplary case, only payload can be used. If slots of one path have equal length, the in¬ formation about paths and slot-lengths can be preconfigured in each intermediate node. Since each node knows at each time which slot comes next, what length the slot has and where to switch to, no additional header information is required. Slot lengths and path sequence:
Optionally, slot-length and path sequence can change over time or with each cycle. The change however, can be configu- red in advance.
Fig.5 shows a portion of a network comprising a single node 501 with three ports I, II and III as well as a switching table 502 and a switching table 503. Table 502 shows an exemplary switching for the port I with fixed slot lengths and sequences, whereas table 503 shows an exemplary switching for the port I with a change of slot lengths and changing se¬ quences. Also, in table 503, the duration of the slots can be adapted .
Configuration of flexible TDM paths:
A duration of the flexible slots, a path sequence and a route can be configured via a centralized or via a decentralized approach.
The centralized approach can be performed, e.g., by a Network Management System (NMS) . As a central entity, the NMS may be aware of traffic requests as well as the network topology. After calculating the routes for each traffic (or portion thereof) , the NMS assigns slot durations and sequences for each path. Each node can be configured by the NMS via ap¬ propriate protocols, e.g., by using SNMP . The decentralized approach can be performed similar to an ap¬ proach configuring MPLS paths with RSVP-TE: When a node decides to establish or to change a path, a reservation message can be sent from the source node to the destination node. The path can be routed at the source or information may be sent hop-by-hop to the destination node. The reservation request message can include information about the path characte¬ ristics, e.g., required bandwidth, maximum delay, etc. When all nodes along the path are provided with the required re- sources, a confirmation message is sent in backward directi¬ on. The confirmation message triggers the actual reservation of bandwidth. Additionally, slot lengths and preferred slot times can be proposed in the response message. Upstream nodes can configure their switching behavior accordingly. It is noted that the slot lengths and path sequences may have local link characteristics or can be identical for the whole path.
Slot in Slot:
If required, slots can be sub-divided to carry traffic of smaller slots. This mechanism can be used to speed up forwarding of traffic if many small traffic portions are transpor¬ ted via common sub-paths. Hence, an encapsulation and de- capsulation can be conducted at the outer path end-nodes. In¬ formation about the sub-slot structure can be included in the outer slot header or can be preconfigured in the end-nodes.
Fig.6 shows an outer slot 601 with a header and several inner slots 602 to 606, each comprising a header of its own.
Further Advantages:
The approach suggested herein combines advantages of packet and TDM switched networks, in particular:
- No fixed TDM slots are required; hence, an available capacity can be efficiently utilized.
- Time slots can be freely adjusted.
- Frame or container handling is simplified as there is no need for deep packet inspection to determine a destination information.
- Quality of service characteristics can be guaranteed and predetermined for all traffic.
- Costs for the nodes can be reduced, because of a re- duced requirement regarding a buffer size (less need for fast memory) ; in addition, time-slots can be assigned in order to efficiently utilize available buffers . Hence, compared to current TDM networks using fixed-sized containers, the approach suggested allows for a higher degree of flexibility as any granularity of bitrates can be selec- ted, e.g., 893827 bps or 1 Gbps . This further allows for a less complex operation without any need for mapping of client data to a fixed container size. Also, as the overhead is significantly reduced, this approach is bandwidth efficient. ODU traffic can be transported transparently by the approach provided. In particular, the approach suggested may utilize an optical transport layer for conveying traffic. Hence, dif¬ ferent clients may co-exist and nodes of the network may pro¬ vide services of ODU clients. This is in particular useful for an end-to-end relation in case both end nodes support ODU traffic and flexible TDM traffic as described herein.
A remapping of ODU traffic into any slot-size of the flexible TDM suggested can be done at intermediate nodes if required (e.g. for using multipath) . The same ODU format may be reestablished at the other side of the flexible TDM domain.
List of Abbreviations:
MPLS Multiprotocol Label Switching
NMS Network Management System
ODU Optical Data Unit
OTH Optical Transport Hierarchy
OTN Optical Transport Network
RSVP-TE Resource Reservation Protocol - Traffic Engineering
SDH Synchronous Digital Hierarchy
SONET Synchronous Optical Network
STM Synchronous Transport Module
TDM Time Division Multiplex
VC Virtual Channel
VCAT Virtual Concatenation

Claims

A method for processing data in a network element,
- wherein the network element provides at least one
outgoing data stream, wherein said outgoing data stream comprises transport slots,
- wherein the transport slots of the at least one out¬ going data stream are configured according to a flexible bitrate.
The method according to claim 1, wherein said transport slots are TDM transport slots.
The method according to any of the preceding claims, wherein the transport slots of the at least one outgoing data stream are configured according to a flexible bi¬ trate at least partially based on at least one of the following criteria:
- to guarantee a requested bitrate for a particular
service ;
- to provide a reduced end-to-end delay;
- to ensure reduced transport costs.
The method according to any of the preceding claims, wherein the transport slots of the at least one outgoing data stream are configured based on a preconfigured, an¬ ticipated or predicted time-slot assignment.
The method according to claim 4, wherein the at least one outgoing data stream is configured via a network management system.
The method according to any of the preceding claims, wherein the transport slots of the at least one outgoing data stream are configured based on a reactive time-slot assignment .
The method according to claim 6, wherein at least one incoming data packet comprises information regarding a time slot assignment to be utilized by the network ele¬ ment .
8. The method according to any of claims 6 or 7, wherein the transport slots of the at least one outgoing data stream are configured according to a flexible bitrate based on an order of incoming data packets and an as¬ signment of the incoming data packets to outgoing data streams, wherein such assignment is conveyed with the data packets.
9. The method according to any of the preceding claims, wherein the network element is a node of a network, in particular a node of at least one communication path.
10. The method according to any of the preceding claims, wherein the network element conveys data of a communica¬ tion via several transport slots of at least two logical channels .
11. The method according to any of the preceding claims, wherein the network element conveys data of a communica¬ tion via separate outgoing data streams.
12. A device comprising or being associated with a process¬ ing unit that is arranged such that the method according to any of the preceding claims is executable thereon.
13. The device according to claim 12, wherein the device is a network element, in particular a node of a communica¬ tion network, wherein said node is in particular a portion of at least one communication path.
14. A communication system comprising at least one device according to claim 13.
EP09749074A 2009-11-02 2009-11-02 Method and device for processing data in a network element Withdrawn EP2497213A1 (en)

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/EP2009/064450 WO2011050860A1 (en) 2009-11-02 2009-11-02 Method and device for processing data in a network element

Publications (1)

Publication Number Publication Date
EP2497213A1 true EP2497213A1 (en) 2012-09-12

Family

ID=42237417

Family Applications (1)

Application Number Title Priority Date Filing Date
EP09749074A Withdrawn EP2497213A1 (en) 2009-11-02 2009-11-02 Method and device for processing data in a network element

Country Status (2)

Country Link
EP (1) EP2497213A1 (en)
WO (1) WO2011050860A1 (en)

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20040028071A1 (en) * 1999-09-10 2004-02-12 Interval Research Corporation, A Washington Corporation Apparatus and method for managing variable-sized data slots with timestamp counters within a TDMA frame
US6920113B1 (en) * 2000-03-28 2005-07-19 Telsima Inc. Transport of iscochronous and bursty data on a sonet ring
JP3916953B2 (en) * 2001-12-28 2007-05-23 日本テキサス・インスツルメンツ株式会社 Variable time division multiplexing transmission system

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See references of WO2011050860A1 *

Also Published As

Publication number Publication date
WO2011050860A1 (en) 2011-05-05

Similar Documents

Publication Publication Date Title
US7075927B2 (en) Method and system for quality of service (QoS) support in a packet-switched network
US6970424B2 (en) Method and apparatus to minimize congestion in a packet switched network
Qiao Labeled optical burst switching for IP-over-WDM integration
US6693909B1 (en) Method and system for transporting traffic in a packet-switched network
US6654374B1 (en) Method and apparatus to reduce Jitter in packet switched networks
US6714517B1 (en) Method and apparatus for interconnection of packet switches with guaranteed bandwidth
US6320863B1 (en) Backplane architecture for dynamic synchronous transfer mode
US6647428B1 (en) Architecture for transport of multiple services in connectionless packet-based communication networks
US7583661B2 (en) Method and apparatus for improved IP networks and high-quality services
US7649910B1 (en) Clock synchronization and distribution over a legacy optical Ethernet network
US6771663B1 (en) Hybrid data transport scheme over optical networks
US8416770B2 (en) Universal service transport transitional encoding
US20040042495A1 (en) Dynamically channelizable packet transport network
US20040202148A1 (en) System and method of data stream transmission over MPLS
US9832107B2 (en) Misconnection avoidance on networks
US20030026250A1 (en) Method and device for synchronous cell transfer and circuit-packet duality switching
US6973084B1 (en) Hybrid data transport scheme over optical networks
CN101212424A (en) Ethernet switching method and device combining circuit switching and packet switching
JP2006513672A (en) Periodic optical packet switching
US9037746B2 (en) Pseudowire circuit emulation
WO2011050860A1 (en) Method and device for processing data in a network element
US7813375B2 (en) Process to transfer a time division multiplexing (TDM) frame over a MPLS network
Sheeshia et al. Supporting Ethernet in optical-burst-switched networks
Kim et al. The role of Ethernet over SDH in QoS switch/router
Waqar et al. QoS assurance for PON-based fronthaul and backhaul systems of 5G cloud radio access networks

Legal Events

Date Code Title Description
PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

17P Request for examination filed

Effective date: 20120604

AK Designated contracting states

Kind code of ref document: A1

Designated state(s): AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO SE SI SK SM TR

DAX Request for extension of the european patent (deleted)
RAP1 Party data changed (applicant data changed or rights of an application transferred)

Owner name: NOKIA SOLUTIONS AND NETWORKS OY

RAP1 Party data changed (applicant data changed or rights of an application transferred)

Owner name: XIEON NETWORKS S.A.R.L.

RAP1 Party data changed (applicant data changed or rights of an application transferred)

Owner name: XIEON NETWORKS S.A R.L.

17Q First examination report despatched

Effective date: 20150604

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE APPLICATION IS DEEMED TO BE WITHDRAWN

18D Application deemed to be withdrawn

Effective date: 20151015