Switch based network architecture for IP multicast and integrated services

• US 6,671,276 B1
• Filed: 11/18/1997
• Issued: 12/30/2003
• Est. Priority Date: 11/18/1997
• Status: Expired due to Term

• Current Assignee: Juniper Networks Incorporated
• Original Assignee: NEC Corporation
• Inventors: Nishida, Takeshi, Bakre, Ajay V.
• Primary Examiner(s): Kizou, Hassan
• Assistant Examiner(s): Tsegaye, Saba
• Application Number:
  US08/972,748
• Time in Patent Office: 2,233 Days
• Field of Search: 370/395.L, 370/390, 370/432, 370/396, 370/397, 370/398, 370/399, 370/400, 370/401, 370/409, 370/352, 370/353, 370/354, 370/355, 370/905, 370/395.5, 709/224, 709/225, 709/226, 709/227, 709/228, 709/229
• US Class Current: 370/395.1
• CPC Class Codes: H04L 12/1836 with heterogeneous network architecture
• CPC Class Codes: H04L 12/4608 LAN interconnection over ATM networks
• CPC Class Codes: H04L 2012/5621 Virtual private network [VPN]; Private-network - network-interface (P-NNI)
• CPC Class Codes: H04L 2012/563 Signalling, e.g. protocols, reference model
• CPC Class Codes: H04L 2012/5642 Multicast/broadcast/point-multipoint, e.g. VOD
• CPC Class Codes: H04L 2012/5667 IP over ATM
• CPC Class Codes: H04L 47/70 Admission control; Resource allocation
• CPC Class Codes: H04L 47/724 at intermediate nodes, e.g. resource reservation protocol [RSVP]
• CPC Class Codes: H04L 47/786 Mapping reservation between domains
• CPC Class Codes: H04L 47/806 Broadcast or multicast traffic
• CPC Class Codes: H04L 47/827 Aggregation of resource allocation or reservation requests
• CPC Class Codes: H04Q 11/0478 Provisions for broadband connections

Abstract

A method and a network architecture for mapping Internetworking Protocol (IP) multicast and Integrated Services (i.e. differing levels of quality of service) over asynchronous transfer mode (ATM) networks, based upon multicast switches, allowing IP/resource reservation protocol (RSVP) applications running on ATM hosts to seamlessly participate in Internet-wide multicast sessions. The method and architecture employ ATM capabilities to support features such as receiver heterogeneity, shortcut routing and scalability.

Claims

1. A method for mapping quality of service based internetworking protocol (IP) multicast in a network architecture, said IP multicast supporting at least one multicast group, said network architecture comprising an asynchronous transfer mode (ATM) cloud including a plurality of logical IP subnets, a plurality of multicast switches and a plurality of local ATM hosts, said multicast switches communicating using ATM protocols, wherein one of said plurality of multicast switches is located in each of said logical IP subnets and at least one of said local ATM hosts is located in each of said logical IP subnets, said method comprising the steps of:

forming an intra logical IP subnet control tree in each of said logical IP subnets for communication between said multicast switches and said ATM hosts;

forming an inter logical IP subnet control tree for providing shortcut routing for inter logical IP subnet multicast traffic; and

for each of said at least one multicast group,

forming a best effort point to point data virtual circuit between one of said ATM hosts sending multicast data and one of said multicast switches located in one of said logical IP subnets in which said one of said ATM hosts sending multicast data is disposed;

forming a best effort intra logical IP subnet data tree in each of said logical IP subnets;

forming a best effort inter logical IP subnet data tree, wherein said best effort inter logical IP subnet data tree may be concatenated with said best effort intra logical IP subnet data trees, thereby forming shortcuts through said multicast switches for routing inter logical IP subnet multicast traffic;

upgrading at least one branch of at least one of said best effort intra logical IP subnet data trees in said logical IP subnets to be a quality of service intra logical IP subnet data tree for transmitting multicast data;

upgrading said best effort inter logical IP subnet data tree to a quality of service inter logical IP subnet data tree for transmitting multicast data;

concatenating said quality of service intra logical IP subnet data trees with said quality of service inter logical IP subnet data tree and any remaining of said best effort intra logical IP subnet data trees, thereby forming a quality of service based multicast tree.

2. The method according to claim 1, wherein said step of forming an intra logical IP subnet control tree in each of said logical IP subnets comprises:

establishing point to multipoint control virtual circuits within each of said logical IP subnets using ATM signaling, between said multicast switches and said local ATM hosts disposed in each of said logical IP subnets.

3. The method according to claim 1, wherein said step of forming an inter logical IP subnet control tree further comprises:

establishing point to multipoint control virtual circuits between said multicast switches located in each of said logical IP subnets using ATM signaling.

4. The method according to claim 3, further comprising the steps of:

each of said multicast switches determining, using group membership information regarding said at least one multicast group, a specific set of others of said multicast switches needing to receive multicast data originating from any of said local ATM hosts sending multicast data for each of said at least one multicast group.

5. The method according to claim 1, wherein said step of forming a best effort intra logical IP subnet data tree comprises:

determining sets of said local ATM hosts in said logical IP subnets as being members of one of said at least one multicast group;

establishing best effort point to multipoint data virtual circuits within each of said logical IP subnets for each of said at least one multicast group using ATM signaling from a respective one of said multicast switches in each of said logical IP subnets to each of said local ATM hosts which are determined to be members of said one of said at least one multicast group.

6. The method according to claim 5, wherein said ATM cloud further comprises a plurality of multicast address resolution servers, at least one of said multicast address resolution servers being located in each of said logical IP subnets, said step of determining sets further comprising the steps of:

registering ones of said ATM hosts receiving multicast data with said multicast address resolution servers when said ones of said ATM hosts receiving multicast data wish to join one of said at least one multicast group, said multicast address resolution servers resolving an IP multicast address to an ATM address for each of said ones of said ATM hosts receiving multicast data, and

registering said multicast switches with said multicast address resolution servers as promiscuous receivers and as multicast servers for all IP multicast addresses.

7. The method according to claim 6, wherein one of said ATM hosts sending multicast data registers as a sender with one of said multicast address resolution servers located in one of said logical IP subnets containing said sender, said sender providing said sender'"'"'s ATM address and one of said IP multicast addresses to which said sender wishes to send said multicast data, said one of said multicast address resolution servers returning an ATM address of said one of said local multicast switches located in said one of said logical IP subnets containing said sender as a sole receiver of said one of said at least one multicast group.

8. The method according to claim 7, wherein additional ones of said ATM hosts sending multicast data begin sending multicast data to one of said at least one multicast group, thereby allowing said one of said ATM hosts sending multicast data and said additional ones of said ATM hosts sending multicast data to send said multicast data to a single IP multicast group address at a given time.

9. The method according to claim 6, wherein changes in membership of said one of said at least one multicast group are communicated to said multicast switches by said multicast address resolution servers, said multicast switches adding and removing virtual circuits to said ones of said ATM hosts receiving multicast data according to said changes in membership of said one of said at least one multicast group.

10. The method according to claim 1, wherein said step of forming a best effort inter logical IP subnet data tree further comprises:

forwarding group membership information regarding said at least one multicast group from each of said multicast switches regarding its logical IP subnet to others of said multicast switches;

each of said multicast switches determining, using said group membership information, a specific set of others of said multicast switches needing to receive multicast data originating from any of said local ATM hosts sending multicast data; and

establishing a best effort point to multipoint data virtual circuit from said one of said multicast switches disposed in one of said logical IP subnets containing one of said ATM hosts sending multicast data to others of said multicast switches disposed in said logical IP subnets containing ones of said ATM hosts receiving multicast data using ATM signaling.

11. The method according to claim 1, further comprising the steps of:

after said step of forming a best effort inter logical IP subnet data tree, transmitting a source description control message from said one of said ATM hosts sending multicast data, over a first point to point control virtual circuit established between said one of said ATM hosts sending multicast data and said one of said multicast switches located in said one of said logical IP subnets in which said one of said ATM hosts sending multicast data is disposed;

forwarding said source description control message from said multicast switch in said logical IP subnet of said one of said ATM hosts sending multicast data to others of said multicast switches over said inter logical IP subnet control tree; and

said others of said multicast switches forwarding said source description control message over said intra logical IP subnet control tree to others of said ATM hosts receiving multicast data.

12. The method according to claim 11, further comprising the steps of:

after said step of others of said multicast switches forwarding said source description control message, said multicast switches receiving quality of service resource reservation request control messages over second ones of point to point control virtual circuits, from ones of said ATM hosts receiving multicast data and desiring quality of service based multicast in said logical IP subnets, wherein each of said messages is capable of indicating a different quality of service, said multicast switches aggregating said quality of service resource reservation request control messages, forming aggregate quality of service resource reservation request control messages, said quality of service resource reservation request control messages indicating resource requirements from said ones of said ATM hosts receiving multicast data; and

said multicast switches transmitting said aggregate quality of service resource reservation request control messages over third ones of point to point control virtual circuits to said multicast switch in said logical IP subnet of said one of said ATM hosts sending multicast data, said third ones point to point control virtual circuits being formed for transmitting said aggregate quality of service resource reservation control messages, said one of said multicast switches aggregating and forwarding said quality of service resource reservation request control messages to said one of said ATM hosts sending data, over said first point to point control virtual circuit.

13. The method according to claim 12, wherein said quality of service resource reservation control messages are used in said step of upgrading said at least one branch of at least one of said best effort intra logical IP subnet data trees.

14. The method according to claim 12, further comprising the step of:

upgrading said best effort point to point data virtual circuit to a quality of service point to point data virtual circuit, wherein said aggregate quality of service resource reservation control messages are used in said step of upgrading said best effort inter logical IP subnet data tree and said step of upgrading said best effort point to point data virtual circuit.

15. The method according to claim 14, wherein quality of service parameters of said quality of service inter logical IP subnet data trees are sufficiently large to accommodate one of said quality of service resource reservation request control messages having a largest quality of service requirement.

16. The method according to claim 12, further comprising the steps of:

transmitting new quality of service resource reservation request control messages from ones of said receivers to respective ones of said multicast switches to change said quality of service based multicast tree;

changing at least some branches of said quality of service intra logical IP subnet data trees to new quality of service intra logical IP subnet data trees for transmitting multicast data using ATM signaling, said upgrading being based on said new quality of service resource reservation request control messages; and

concatenating said new quality of service intra logical IP subnet data trees with said quality of service intra logical IP subnet data trees, said quality of service inter logical IP subnet tree and said any remaining of said best effort intra logical IP subnet data trees, thereby forming a new quality of service based multicast tree.

17. The method according to claim 12, further comprising the step of detecting an absence of said source description control message and said quality of service resource reservation request control messages for a predetermined period of time and deleting corresponding ones of said quality of service intra IP logical subnet data trees after said predetermined period of time has elapsed.

18. The method according to claim 12, wherein said quality of service resource reservation request control messages generated by said ones of said ATM hosts receiving multicast data, further include a filter, said filter indicating said quality of service resource reservation request control messages apply to data sent by one of all current and future ones of said senders, a specific list of said senders and a fixed one of said senders.

19. The method according to claim 18, wherein when said filter indicates that said quality of service resource reservation request control messages apply to a fixed one of said senders, a separate quality of service based multicast tree is formed for said fixed one of said senders.

20. The method according to claim 18, wherein additional intra logical IP subnet quality of service based multicast trees are formed for each of said additional ones of said ATM hosts sending multicast data, said receivers being added to said additional intra logical IP subnet quality of service based multicast trees based upon said filter.

21. The method according to claim 18, wherein said receivers are partitioned into groups, said groups grouping ones of said receivers based upon same ones of said quality of service resource reservation request control messages and same ones of said filter, each of said groups operating on a separate multicast tree.

22. The method according to claim 1, wherein after said step of upgrading at least one branch of at least one of said best effort intra IP logical subnet trees to quality of service intra logical IP subnet data trees, upgraded ones of said at least one branch are removed.

23. The method according to claim 1, further comprising interoperating with protocols outside of said ATM cloud through edge switches, said edge switches being multicast switches and disposed at edges of said ATM cloud.

24. The method according to claim 23, wherein said edge switches forward group information learned from outside networks to said multicast switches in said ATM cloud.

25. The method according to claim 24, wherein said edge switches cooperate in determining a given one of said edge switches to be associated with a given outside sender disposed outside said ATM cloud, said given one of said edge switches initiating inter logical IP subnet virtual circuits and intra logical IP subnet virtual circuits in said ATM cloud for said given outside sender.

26. A network architecture for mapping quality of service based internetworking protocol (IP) multicast in a network, said IP multicast supporting at least one multicast group, said network architecture comprising an asynchronous transfer mode (ATM) cloud, said ATM cloud comprising:

a plurality of logical IP subnets;

a plurality of ATM hosts, wherein at least one of said plurality of ATM hosts is disposed in each of said logical IP subnets;

a plurality of multicast switches, wherein each of said plurality of multicast switches is disposed in a respective one of said logical IP subnets, for aggregating senders outside each of said logical IP subnets for local receivers inside each of said logical IP subnets and aggregating receivers outside each of said logical IP subnets for local senders inside each of said logical IP subnets, said multicast switches communicating with each other using ATM protocols; and

a plurality of multicast address resolution servers, wherein one of said plurality of multicast address resolution servers is located in each of said logical IP subnets for resolving an IP multicast address to ATM addresses of receivers joining one of said at least one multicast group represented by a multicast address;

further comprising edge switches for aggregating ones of said ATM hosts receiving multicast data inside said ATM cloud, for senders outside said ATM cloud and aggregating receivers outside said ATM cloud for ones of said ATM hosts sending multicast data inside said ATM cloud, said edge switches being ones of said multicast switches disposed at edges of said ATM cloud.

27. A network architecture according to claim 26, wherein said edge switches interoperate with inter domain multicast routing (IDMR) protocols used outside said ATM cloud for sending and receiving multicast route and membership information about said ATM cloud.

28. A network architecture according to claim 27, wherein for each of said senders outside said ATM cloud, only one of said edge switches establishes a multicast tree within said ATM cloud.

29. In a network architecture for mapping internetworking protocol (IP) multicast and integrated services over an asynchronous transfer mode (ATM) network, said network architecture comprising an ATM cloud including a plurality of multicast switches and local ATM hosts, said ATM cloud further including a plurality of logical IP subnets, each of said logical IP subnets containing one of said multicast switches and at least one of said local ATM hosts, said multicast switches communicating with each other using ATM protocols, wherein each of said multicast switches comprises:

switching fabric allowing incoming virtual circuits to feed multiple outgoing virtual circuits;

a switch controller for controlling said switching fabric to establish header translation tables for said virtual circuits;

an ATM signaling software module for establishing said virtual circuits using ATM signaling, said switch controller being capable of terminating said virtual circuits and concatenating said virtual circuits on two different links, thereby allowing data on one of said incoming virtual circuits to be forwarded by said switching fabric on one of said outgoing virtual circuits without software intervention; and

a multicast routing module, including a module for maintaining group membership information for said logical IP subnets, and a module for communicating with peer functions on said multicast switches in said ATM cloud for exchanging summarized local membership information.

30. A network architecture according to claim 29, wherein said multicast switches further comprise:

a resource reservation protocol message handling module for terminating resource reservation control messages from said multicast switches in said network architecture and from said local ATM hosts;

a call admission control module connected to said resource reservation protocol message handling module for determining if sufficient resources may be reserved for a new data flow when said resource reservation protocol message handling module receives a resource reservation for said new data flow; and

a virtual circuit management module for establishing virtual circuits for said new flow when said call admission control module determines sufficient resources may be reserved.

31. A network architecture according to claim 30, wherein said virtual circuit management module comprises a first part for establishing intra logical IP subnet virtual circuits and inter logical IP subnet virtual circuits and a second part for concatenating any two of said intra logical IP subnet virtual circuits and said inter logical IP subnet virtual circuits into one virtual circuit, said two virtual circuits terminating in a respective one of said multicast switches, said intra logical IP subnet virtual circuits and inter logical IP subnet virtual circuits forming a multicast tree among said multicast switches.

32. A network architecture according to claim 31, wherein said multicast tree is one of a best effort based multicast tree, a quality of service based multicast tree or a mixed best effort and quality of service based multicast tree.

33. A network architecture according to claim 31, wherein said virtual circuit management module adds ones of said ATM hosts receiving multicast data as leaf nodes to said multicast tree when said ones of said ATM hosts receiving multicast data join said multicast tree.

34. A network architecture according to claim 33, wherein said virtual circuit management module adds new ones of said multicast switches to said multicast tree when said new ATM hosts join said multicast tree.

35. A network architecture according to claim 31, wherein said virtual circuit management module removes ones of said ATM hosts receiving multicast data as leaf nodes from said multicast tree when said ones of said ATM hosts receiving multicast data leave a data flow.

36. A network architecture according to claim 35, wherein said virtual circuit management module removes ones of said multicast switches from said multicast tree when all ones of said ATM hosts receiving multicast data disposed in corresponding ones of said logical IP subnets containing said ones of said multicast switches are removed.

37. A network architecture according to claim 30, wherein at least one of said multicast switches is an edge switch for aggregating ones of said ATM hosts receiving multicast data inside said ATM cloud for hosts being senders outside said ATM cloud, and aggregating hosts being receivers outside said ATM cloud for ones of said ATM hosts sending multicast data inside said ATM cloud.

38. A network architecture according to claim 37, wherein said multicast routing module further comprises a module for providing an inter domain multicast routing (IDMR) protocol interface to IP routers located outside said ATM cloud.

39. A network architecture according to claim 38, wherein said inter domain multicast routing (IDMR) protocol interface comprises multicast routing for supported IDMR protocols for interacting with said IP routers outside said ATM cloud by sending and receiving multicast route and membership information about said ATM cloud.