초록 ▼

The invention provides a method and network communication equipment for low latency loss-free burst switching. Burst-transfer schedules are determined by controllers of bufferless core nodes according to specified bitrate allocations and distributed to respective edge nodes. In a composite-star netw

The invention provides a method and network communication equipment for low latency loss-free burst switching. Burst-transfer schedules are determined by controllers of bufferless core nodes according to specified bitrate allocations and distributed to respective edge nodes. In a composite-star network, burst schedules are initiated by any core node. Burst formation takes place at source edge nodes and a permissible burst size is determined according to an allocated bitrate of a burst stream to which the burst belongs. The permissible burst size is subject to constraints such as permissible burst-formation delay, a minimum guard-time requirement, and permissible delay jitter. A method of control-burst exchange between each edge node and each bufferless core node enables burst scheduling, time coordination, and loss-free burst switching. Both the payload bursts and control bursts are carried by optical channels connecting the edge nodes and the core notes.

대표청구항 ▼

1. A method of burst switching in a communication network having a plurality of core switching nodes and a plurality of edge nodes, each core switching node comprising at least one optical switch having no traffic buffers, the method comprising: receiving, from an edge node at a selected core switch

1. A method of burst switching in a communication network having a plurality of core switching nodes and a plurality of edge nodes, each core switching node comprising at least one optical switch having no traffic buffers, the method comprising: receiving, from an edge node at a selected core switching node, a bit rate allocation request specifying a destination edge node;generating, by the selected core switching node in response to the bit allocation request, at least one burst transfer permit, each burst transfer permit specifying a respective permissible burst size, an inter-burst interval and the destination edge node, the respective permissible burst size having: an upper bound constrained by a maximum jitter tolerance of data constituting data bursts received from the edge node; anda lower bound much larger than a time required for reconfiguring the at least one optical switch and a maximum arrival time error for data bursts arriving at the at least one optical switch; andsending the at least one burst transfer permit by the selected core switching node to the edge node. 2. The method of claim 1, further comprising: receiving, at the edge node from the selected core switching node, the at least one burst transfer permit;assembling, at the edge node, at least one data burst having a burst size not exceeding the respective permissible burst size specified in the burst transfer permit; andsending the at least one data burst from the edge node to the selected core switching node. 3. The method of claim 2, wherein: each burst transfer permit specifies an arrival time for a data burst at the selected core switching node; andsending the at least one data burst from the edge node to the selected core switching node comprises timing the sending of a data burst based on the arrival time specified in a corresponding burst transfer permit. 4. The method of claim 1, further comprising: receiving, by the selected core switching node from the edge node, at least one data burst based at least in part on the at least one burst transfer permit; andoptically switching, at the selected core switching node, the at least one data burst to the destination edge node. 5. The method of claim 1, wherein: the data of data bursts switched by the selected core switching node comprises data of at least one service; andthe respective permissible burst size has an upper bound constrained by a delay tolerance of the at least one service switched by the selected core switching node. 6. The method of claim 1, wherein each burst transfer permit specifies an arrival time for a data burst at the selected core switching node. 7. The method of claim 1, further comprising: receiving, from a plurality of edge nodes at each core switching node, a plurality of bit rate allocation requests, each bit rate allocation request specifying a respective destination edge node;generating, by the core switching nodes in response to the bit allocation requests, respective burst transfer permits, each burst transfer permit specifying a respective permissible burst size, an inter-burst interval and a respective destination edge node, the respective permissible burst size having an upper bound constrained by a delay tolerance of services switched by the core switching nodes; andsending the burst transfer permits by the core switching nodes to the edge nodes. 8. The method of claim 7, further comprising: receiving, at the edge nodes from the core switching nodes, respective burst transfer permits; andassembling, at the edge nodes in response to the respective burst transfer permits, respective data bursts having a burst sizes not exceeding the respective permissible burst sizes specified in the burst transfer permits; andsending the respective data bursts from the edge nodes to respective core switching nodes. 9. The method of claim 8, wherein: each burst transfer permit specifies an arrival time for a data burst at the selected core switching node; andsending the respective data bursts from the edge nodes to respective core switching nodes comprises timing the sending of the data bursts based on respective arrival times specified in respective corresponding burst transfer permits. 10. The method of claim 7, further comprising: receiving, by the core switching nodes from the edge nodes, respective data bursts based at least in part on the burst transfer permits; andswitching, at the core switching nodes, the at data bursts to respective destination edge nodes. 11. The method of claim 7, wherein the respective permissible burst sizes have a lower bound much larger than a time required for reconfiguring the optical switches and a maximum arrival time error for data bursts arriving at the optical switches. 12. The method of claim 7, wherein each edge node simultaneously transfers data bursts to plural core switching nodes. 13. The method of claim 12, wherein the communication network has co-located edge nodes and core switching nodes, comprising equalizing propagation delays from the edge nodes to the core switching nodes. 14. The method of claim 12, further comprising time locking edge nodes to respective core switching nodes. 15. The method of claim 7, further comprising: associating data bursts with respective burst streams; andsizing the data bursts based at least in part on at least one attribute of the burst stream. 16. The method of claim 15, wherein the at least one attribute of the burst stream comprises a service class. 17. The method of claim 7, further comprising: associating data bursts with respective burst streams; andswitching all data bursts of each burst stream in a respective core switching node. 18. The method of claim 1, wherein the bit rate allocation request specifies a requisite bit rate. 19. The method of claim 18, wherein the selected core switching node determines an update bit rate to replace the requisite bit rate. 20. The method of claim 1, wherein the core switching nodes are arranged in a composite star configuration.