Multiplex communications – Pathfinding or routing – Switching a message which includes an address header
Reexamination Certificate
1998-08-13
2002-03-26
Chin, Wellington (Department: 2664)
Multiplex communications
Pathfinding or routing
Switching a message which includes an address header
C370S400000
Reexamination Certificate
active
06363072
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a communication control system which conducts IP (Internet Protocol) packet distribution over an ATM (Asynchronous Transfer Mode) network and a communication control method thereof and, more particularly, to an ATM network communication control system attaining high-speed transmission through reduction of such packet transmission by hop-by-hop such as that by a router within a network, and a communication control method thereof.
2. Description of the Related Art
Among conventional art of this kind for conducting IP (Internet Protocol) packet distribution over an ATM network, the following four are well known.
Classical IP model and Tag Switching
MPOA
IP Switch
Cell Switch Router
Characteristics of the respective art will be described in the following.
a) Characteristics of Classical IP model and Tag Switching
FIG. 5
is a schematic diagram showing structure of an ATM network to which the classical IP model and a tag switching are applied. For communication between different IP sub-nets according to this art, a router is indispensable between the IP sub-nets as shown in FIG.
5
.
Classical IP model and tag switching, however, have a drawback that no improvement can be expected of a transmission rate in communication between different IP sub-nets. The reason is that a router always intervening between IP sub-nets to conduct packet transmission by software processing bottlenecks packet transmission.
b) Characteristics of MPOA
FIG. 6
is a schematic diagram showing structure of an ATM network to which MPOA is applied. As illustrated in
FIG. 6
, an MPC (Multiprotocol Client) and an MPS (Multiprotocol Server) are provided in the network. This system employs LANE (LAN Emulation, LAN represents Local Area Network) and NHRP (Next Hop Resolution Protocol) techniques. Within the same IP subnet, an LANE protocol is used for communication. For communication between different IP sub-nets, an MPC first serves as a router and then a direct connection is set up between MPSs of a transmission source and a transmission destination to carry out communication.
In
FIG. 6
, for communication from an MPS
1
to an MPS
2
, an MPC sets up a shortcut VC (Virtual Channel) to each of the MPS
1
and the MPS
2
. At this state, the MPC serves as a router to carry out communication. If at this occasion, communication to the same destination is conducted frequently, an NHRP protocol operates, so that the MPC
1
sets up a shortcut VC to the MPS
2
. At this state, direct communication is possible between the MPS
1
and the MPS
2
to enable high-speed transfer of packets.
MPOA, however, has a drawback that efficiency is not high because an end-to-end connection can be set up only after connections are established between adjacent MPC and MPS and between adjacent MPSs. The reason is that an end-to-end connection is enabled after the MPS sends or receives a protocol to/from its adjacent node.
Another drawback is that when a new packet is generated, load on an MPS in processing will be sharply increased. The reason is that when a new packet is generated, each node conducts address resolution and creation of a table by sending or receiving a protocol to/from its adjacent node.
A further drawback is that as a network is increased in scale, address resolution resources of the entire network are exponentially increased. The reason is that as the network is increased in scale, the number of entries in an address resolution table which the MPS has as well as the number of MPSs increases.
c) Characteristics of IP Switch
FIG. 7
is a schematic diagram showing structure of an ATM network to which an IP switch is applied. As illustrated in
FIG. 7
, an IP switch gateway (hereinafter abbreviated as GW) and an IP switch (hereinafter abbreviated as IPSW) are provided in the network. A GW mainly conducts termination of an IP switch network. An IPSW is mainly composed of a controller unit (hereinafter abbreviated as CTL) for forwarding an IP packet and a switch unit (hereinafter abbreviated as SW) for switching an ATM cell. A GW is connected to only one IPSW and belongs to a plurality of existing LANs. An IPSW is connected to a plurality of GWs and IPSWs. An IP sub-net provided between a certain IPSW and another IPSW has only these two nodes. Each port of the GW and the IPSW has an IP address.
As illustrated in
FIG. 7
, VCs are always being set up between adjacent devices. On this occasion, a CTL serves as a router to carry out communication by hop-by-hop. If communication to the same destination is conducted frequently, each node used in the communication sends an upstream node a request for preparing a dedicated VC. When there is no abnormality in particular, a dedicated VC will be set up between adjacent nodes. A specific protocol also causes a dedicated VC to be set up between adjacent nodes. When dedicated VCs are established both upstream and downstream the IPSW, the IPSW directly connects the VCs by an SW. In other words, the CTL is prevented from intervening in a communication path. Since at this state, no routing is conducted within the IPSW, packets can be transferred at a high speed.
Using an IP switch, however, has a drawback that efficiency is not high because an end-to-end connection can be set up only after dedicated VCs are established between adjacent GW and IPSW and between adjacent IPSWs. The reason is that an end-to-end connection is enabled after the IPSW sends or receives a protocol to/from its adjacent node.
Another drawback is a total lack of affinity with an approach of an ATM forum. The reason is that a unique protocol is adopted.
A further drawback is that an IP domain is wastefully consumed. The reason is that a port of an IPSW requires an IP address for terminating an IP protocol.
Still further drawback is that a routing protocol should be mounted on every ATM switch. In other words, an ATM network should be composed of IPSWs. The reason is that no communication is possible unless the IPSW conducts routing processing.
Still further drawback is that when a new packet is generated, load on an IPSW in processing will be sharply increased. The reason is that when a new packet is generated, each node conducts address resolution and creation of a table by sending or receiving a protocol to/from its adjacent node.
Still further drawback is that as a network is increased in scale, address resolution resources of the entire network are exponentially increased. The reason is that increase in scale of a network is followed by an increase in the number of entries in an address resolution table which the IPSW has as well as the number of the IPSWs.
d) Characteristics of Cell Switch Router
A cell switch router (hereinafter abbreviated as CSR) is provided in a network. A CSR is mainly composed of a CTL for forwarding an IP packet and an SW for switching an ATM cell. The CSR also conducts termination of a CSR network.
Basic operation of the cell switch router is similar to that of an IP switch. With a VC set up in advance between adjacent devices, the CTL carries out communication by hop-by-hop as a router does. For a specific packet, a dedicated VC is prepared between adjacent nodes which are to be connected by an SW. In other words, the CTL is prevented from intervening in a communication path. Since at this state, no routing is conducted within the CSR, packets can be transferred at a high speed. Protocol used is different from that used for an IP switch.
Conventional IP packet distribution techniques using the above-described cell switch router are recited, for example, in the literature “Cell Switch Router—Basic Concept and Migration Scenario” (Yasuhiro Katsube, Ken-ichi Nagami, Shigeo Matsuzawa, THE INSTITUTE OF ELECTRONICS, INFORMATION AND COMMUNICATION ENGINEERS, TECHNICAL REPORT OF IEICE. SSE95-201, IN95-145 (1996-03)) and the literature “Cell Switch Router—Protocol Mechanism and Implementation Architecture” (Shigeo Matsuzawa, Ken-ichi Nagami, Akiyoshi Mogi, Tatsuya Jinmei, Yasuhiro Katsube, THE INSTITUTE OF ELECTRONICS,
Chin Wellington
NEC Corporation
Pham Brenda H.
Whitham, Curtis & Christonfferosn, PC
LandOfFree
ATM network communication control system allowing end-to-end... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with ATM network communication control system allowing end-to-end..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and ATM network communication control system allowing end-to-end... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2818143