Multiplex communications – Pathfinding or routing – Switching a message which includes an address header
Reexamination Certificate
2000-06-05
2001-08-28
Chin, Wellington (Department: 2664)
Multiplex communications
Pathfinding or routing
Switching a message which includes an address header
C370S389000
Reexamination Certificate
active
06282201
ABSTRACT:
BACKGROUND
The prior art data communication network illustrated in
FIG. 1
includes a plurality of routers such as Local Area Network (LAN) switches having Internet Protocol (IP) routing capability. The LAN switches are connected by a switched network (such as an ATM network) to each other and to LAN emulation clients (LECs) which are directly connected to the ATM network. The LAN switches are connected via ports to several LANs such as Ethernet and/or Token Ring segments. The LECs join emulated LANs (ELANs) and forward data packets to other LECs on the same ELAN over the ATM network. The LAN ports connect to hosts or other communicating stations via, for example, an Ethernet or Token Ring LAN segment.
In
FIG. 1
, three LAN switches A, B and C are connected to an ATM network. LAN switches A and B provide both router and bridge functions while LAN switch C provides a bridge function only. A router D and a host system E (acting as a LEC) are directly connected to the ATM network. LAN switch A has a segment
10
connected to its LAN port. A host system
1
is connected to segment
10
. Another segment
11
is connected to the LAN port on LAN switch B. Two host systems
2
and
4
are connected to segment
11
. Another segment
12
is connected to LAN switch C. A host system
3
is connected to segment
12
.
In
FIG. 1
, host
1
has the IP address A.11 on subnet A. Host
2
has the IP address A.12 on subnet A. Host
3
has the IP address A.13 on subnet A. Host
4
has the IP address B.14 on subnet B. Host E has the IP address B.11 on subnet B. The addresses of the default gateways configured for hosts
1
,
2
,
4
and E are A.2, A.3, B.3 and B.1, respectively. LAN switch A has IP addresses A.2 on subnet A and B.2 on subnet B; its MAC address is M.2. LAN switch B has IP addresses A.3 on subnet A and B.3 on subnet B; its MAC address is M.3. LAN switch C does not have routing capability. Central router D has IP addresses A.1 on subnet A and B.1 on subnet B; its' MAC address is M.1.
Hosts or other stations require a router to communicate with other hosts or stations on IP subnets other than their own. Hosts specify their router with the configured default gateway address. The router function for host
1
is provided by an internal router in LAN switch A. LAN switch B provides the function for hosts
2
and
4
. The router D provides this function for host
3
which is attached to LAN switch C.
When the illustrated network is initially configured the routers are provided with unique Medium Access Control (MAC) addresses and Internet Protocol (IP) addresses (see the assigned addresses described above and illustrated in FIG.
1
). The host systems connected to the segments are configured with their own IP addresses and the address of their default gateway (DGW) for accessing devices not located on the same IP subnet. As long as the network connection remain static this method for operating the network is satisfactory. However, if a device (such as one or more of the host systems) is moved from one segment to another, without reconfiguring the DGW of the device which has moved, an inefficiency is introduced.
For example, if host
2
is moved from segment
11
to segment
10
and needs to communicate with host E, it will forward the packets to its default gateway A.3. The bridge function in LAN switch A will forward the packet over the ATM network to LAN switch B (which is configured with IP address A.3). The router function in LAN switch B will forward the packet via the ATM network to the destination host E. Thus all communications between host
2
and host E will make two trips across the ATM network as opposed to one if host
2
had not been moved from segment
11
to segment
10
.
Two solutions have been proposed to address the problem created when a station such as host
2
moves from one segment served by one router to another served by a different router. Both are unsatisfactory. The first solution is to provide a single central router to serve all hosts or stations on all LAN segments. This solution, however, is not workable (especially if large networks are employed) since it results in a bottleneck at the central router. The other solution is to reconfigure the DGW whenever an end station is moved. This solution is equally unattractive since it imposes a severe administrative problem.
SUMMARY OF THE INVENTION
The invention contemplates a method for configuring a communication network which includes a number of hosts on different subnets, each host being physically connected via a LAN segment to a number of LAN switches. The LAN switches are connected to an ATM network and one or more hosts on one or more subnets are attached directly to the ATM network. All of the LAN switches, which have router function, are configured with at least the same IP address and the same MAC address. The routers in the LAN switches only recognize the common or shared address when they receive an IP ARP from an attached segment. A single router attached to the ATM network could also be configured with the common addresses and responds to this addresses when an IP ARP is received over the ATM network. Hosts attached to the LAN networks are configured with the common IP address as their DGW IP address.
REFERENCES:
patent: 5115433 (1992-05-01), Baran et al.
patent: 5329527 (1994-07-01), Ujihashi et al.
patent: 5357508 (1994-10-01), Le Boudec et al.
patent: 5420862 (1995-05-01), Perlman
patent: 5444702 (1995-08-01), Burnett et al.
patent: 5490139 (1996-02-01), Baker et al.
patent: 5570359 (1996-10-01), Nguyen
patent: 5577033 (1996-11-01), Chang et al.
patent: 5581552 (1996-12-01), Civanlar et al.
patent: 5600644 (1997-02-01), Chang et al.
patent: 5636216 (1997-06-01), Fox et al.
patent: 5636217 (1997-06-01), Moelard
patent: 6091732 (2000-07-01), Alexander, Jr. et al.
IBM Technical Disclosuree Bulletin, v37, #10, 10/94, “Addressing Source Routing in an ATM Emulated Lan”.
IBM Technical Disclosure Bulletin, v39, #08, 08/96, “Method for Improving Network Availability with Redundant Network Servers”.
Alexander, Jr. Cedell Adam
Ervin Jimmy Philip
Lloyd John
Matlack, Jr. Richard Colbert
Vig Deepak
Cesari and McKenna LLP
Chin Wellington
Cisco Technology Inc.
Johnston A. Sidney
Pham Brenda
LandOfFree
Method for configuring distributed internet protocol... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for configuring distributed internet protocol..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for configuring distributed internet protocol... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2465578