公开(公告)号：EP1747644A4

公开(公告)日：2011-09-28

申请号：EP05742868

申请日：2005-04-29

Applicant: IBM

Inventor： LIU ZHEN ,  BACCELLI FRANCOIS ,  CHAINTREAU AUGUSTIN ,  RIABOV ANTON

IPC: H04L12/56 ,  H04L12/18 ,  H04L12/28

CPC classification number: H04L47/263 ,  H04L12/1868 ,  H04L47/10 ,  H04L47/15

公开(公告)号：WO2005109772A2

公开(公告)日：2005-11-17

申请号：PCT/US2005014870

申请日：2005-04-29

Applicant: IBM ,  LIU ZHEN ,  BACCELLI FRANCOIS ,  CHAINTREAU AUGUSTIN ,  RIABOV ANTON

Inventor： LIU ZHEN ,  BACCELLI FRANCOIS ,  CHAINTREAU AUGUSTIN ,  RIABOV ANTON

IPC: H04L12/18 ,  H04L12/28 ,  H04L12/56

CPC classification number: H04L47/263 ,  H04L12/1868 ,  H04L47/10 ,  H04L47/15

Abstract: The present invention addresses scalability and end-to-end reliability in overlay multicast networks. A simple end-system multicast architecture that is both scalable in throughput and reliable in an end-to-end way is used. In this architecture, the transfers between nodes use TCP with backpressure mechanisms to provide data packet transfers between intermediate nodes having finite-size forwarding buffers. There is also a finite­size backup buffer in each node to store copies of packets which are copied out from the receiver window to the forwarding buffers. These backup buffers are used when TCP connections are re-established to supply copies of data packets for the children nodes after their parent node fails, maintaining a complete sequence of data packets to all nodes within the multicast overlay network. The architecture provides end-to-end reliability, tolerates multiple simultaneous node failures and provides positive throughput for any group size and any buffer size.

Abstract translation: 本发明解决了覆盖组播网络中的可扩展性和端到端的可靠性。 使用简单的终端系统多播架构，其吞吐量可扩展，端对端可靠。 在这种架构中，节点之间的传输使用具有背压机制的TCP，以在具有有限大小的转发缓冲器的中间节点之间提供数据分组传输。 每个节点还有一个有利的备份缓冲区，用于存储从接收器窗口复制到转发缓冲区的数据包的副本。 当重新建立TCP连接以在其父节点发生故障之后为子节点提供数据分组的副本时，将使用这些备份缓冲区，将数据包的完整序列维护到组播覆盖网络内的所有节点。 该架构提供端到端的可靠性，可以容忍多个并发节点故障，并为任何组大小和任何缓冲区大小提供正的吞吐量。

公开(公告)号：CA2564363A1

公开(公告)日：2005-11-17

申请号：CA2564363

申请日：2005-04-29

Applicant: IBM

Inventor： LIU ZHEN ,  CHAINTREAU AUGUSTIN ,  BACCELLI FRANCOIS ,  RIABOV ANTON

IPC: H04L12/26 ,  H04L12/18 ,  H04L12/28 ,  H04L12/56

Abstract: The present invention addresses scalability and end-to-end reliability in overlay multicast networks. A simple end-system multicast architecture (Fig. 3) that is both scalable in throughput and reliable in an end-to-end way is used. In this architecture, the transfers between nodes use TCP with backpressure mechanisms to provide data packet transfers between intermediat e nodes (16) having finite-size forwarding buffers (22). There is also a finite~size backup buffer (24) in each node to store copies of packets which are copied out from the receiver window (20) to the forwarding buffers (22). These backup buffers (24) are used when TCP connections are re-established t o supply copies of data packets for the children nodes (18) after their parent node fails, maintaining a complete sequence of data packets to all nodes within the multicast overlay network. The architecture provides end-to-end reliability, tolerates multiple simultaneous node failures and provides positive throughput for any group size and any buffer size.

公开(公告)号：CA2564363C

公开(公告)日：2010-07-20

申请号：CA2564363

申请日：2005-04-29

Applicant: IBM

Inventor： LIU ZHEN ,  BACCELLI FRANCOIS ,  CHAINTREAU AUGUSTIN ,  RIABOV ANTON

IPC: H04L12/26 ,  H04L12/18 ,  H04L12/28 ,  H04L12/56

Abstract: The present invention addresses scalability and end-to-end reliability in overlay multicast networks. A simple end-system multicast architecture (Fig. 3) that is both scalable in throughput and reliable in an end-to-end way is used. In this architecture, the transfers between nodes use TCP with backpressure mechanisms to provide data packet transfers between intermediate nodes (16) having finite-size forwarding buffers (22). There is also a finite~size backup buffer (24) in each node to store copies of packets which are copied out from the receiver window (20) to the forwarding buffers (22). These backup buffers (24) are used when TCP connections are re-established to supply copies of data packets for the children nodes (18) after their parent node fails, maintaining a complete sequence of data packets to all nodes within the multicast overlay network. The architecture provides end-to-end reliability, tolerates multiple simultaneous node failures and provides positive throughput for any group size and any buffer size.