Electrical computers and digital processing systems: multicomput – Computer-to-computer data addressing
Reexamination Certificate
2000-05-05
2003-11-11
Lim, Krisna (Department: 2154)
Electrical computers and digital processing systems: multicomput
Computer-to-computer data addressing
C709S236000, C709S238000, C370S253000, C370S355000, C370S389000, C370S390000
Reexamination Certificate
active
06647428
ABSTRACT:
FIELD OF THE INVENTION
This invention relates to communication networks and, in particular, to a network architecture and corresponding packet header definition used for end-to-end transport through a connectionless packet-based network of multiple services originating at a variety of different types of tributary interfaces.
BACKGROUND
Metropolitan access networks of the future must be able to carry packetized data traffic in a low-cost and efficient manner while at the same time providing a means for transport of synchronous or plesiochronous leased-line services. These networks must be flexible enough to offer a combination of Open Systems Interconnection (OSI) Layer
1
, Layer
2
, and Layer
3
services at lower-speed ingress ports at customer premises, and must be able to uplink aggregated data from these lower-speed interfaces to higher-speed uplink ports located at gateways to core transport networks. These networks must provide facility protection by enabling data to be rerouted in the event of failure, either in mesh or bi-directional ring topologies. These networks must offer different service classes based upon the delay, jitter, and reliability requirements of each customer. These networks must also minimize operational cost by simplifying network management, especially provisioning functions.
Traditional connectionless, packetized data networks based on Internet Protocol (IP) routers or Ethernet switches are inherently closer to meeting the above requirements than connection-oriented asynchronous transfer mode (ATM) networks or time-division-multiplexed synchronous optical network (SONET) rings. However, neither type of network is sufficiently flexible to seamlessly transport data of other protocol types, such as synchronous leased lines, or to offer a combination of OSI Layer
1
,
2
, and
3
services across different ports on a single device. Ethernet as defined in IEEE Standard 802.3-1998 does not have classes of service. Ethernet devices are limited to operation in local area networks (LANs). IP routers do not enable efficient label-based or tag-based lookups at each hop through a network.
Ethernet with a multi-protocol label-switched (MPLS) label stack as defined in the Internet Engineering Task Force (IETF) Internet Draft “MPLS Label Stack Encoding” by E. Rosen et al., draft-ietf-mpls-label-encaps-07.txt, incorporated herein by reference in its entirety, provides 8 classes of service and a 20-bit tag that can be used to switch packets via a label-based lookup at each hop through a network. It also provides an 8-bit time-to-live (TTL) field. Label-switched routers (LSRS) capable of performing a combination of IP routing, MPLS label-based lookups, and policy-based wild-card lookups are able to offer a combination of OSI Layer
1
,
2
, and
3
services across different ports on a single device. The use of MPLS also enables multiple protocol formats encapsulated within Ethernet frames to be carried as interleaved packetized units within the same data stream.
We have taken note of these ideas and built upon them in our architecture for transport of multiple services in connectionless packet-based networks, and in the formulation of the Optical Packet Transport Network (OPTNet) packet format used for data transport in this architecture. (This name is used to refer to the packet format in this specification purely for convenience and is not necessarily the permanent name associated with this packet format.) Part of the OPTNet packet format resembles the Ethernet with MPLS label stack packet format. Other portions of the OPTNet packet format differ to provide additional or different functionality or improved efficiency. Our architecture supports transport of both connectionless packetized data and framed data from synchronous leased lines. In addition, a fundamental premise of our architecture is simplicity in provisioning and management. Our architecture is optimized for OSI Layer
1
and Layer
2
(Virtual LAN, or VLAN) services and for ring topologies, since for these services in a ring it requires no path setup using a label distribution protocol. It supports OSI Layer
1
, Layer
2
, and Layer
3
services.
SUMMARY
An architecture for transport of multiple services in connectionless packet-based networks is described herein, along with the packet format used for data transport in this architecture. The architecture supports transport of both connectionless packetized data and framed data from synchronous leased lines. The architecture supports transparent packetization of incoming DS1 data. The architecture works for mesh architectures but is optimized for OSI Layer
1
(crossconnect) and Layer
2
(Virtual LAN, or VLAN) services and for ring topologies, since for these services in a ring no path setup is required using a label distribution protocol. In addition, it simultaneously supports OSI Layer
1
, Layer
2
, and Layer
3
services.
REFERENCES:
patent: 4901312 (1990-02-01), Hui et al.
patent: 5721819 (1998-02-01), Galles et al.
patent: 5815516 (1998-09-01), Aaker et al.
patent: 5875306 (1999-02-01), Bereiter
patent: 6094525 (2000-07-01), Perlman et al.
patent: 6108338 (2000-08-01), Ramfelt et al.
patent: 6125399 (2000-09-01), Hamilton
patent: 6137797 (2000-10-01), Bass et al.
patent: 6151324 (2000-11-01), Belser et al.
patent: 6208651 (2001-03-01), Van Renesse et al.
patent: 6212183 (2001-04-01), Wilford
patent: 6212185 (2001-04-01), Steeves et al.
patent: 6230144 (2001-05-01), Kilkki et al.
patent: 6289369 (2001-09-01), Sundaresan
patent: 6301254 (2001-10-01), Chan et al.
patent: 6438143 (2002-08-01), Higashida
patent: 6442134 (2002-08-01), Mitchell
patent: 6449279 (2002-09-01), Belser et al.
patent: 6473397 (2002-10-01), Au
patent: 6512768 (2003-01-01), Thomas
patent: 6519737 (2003-02-01), Derby
patent: 2002/0160771 (2002-10-01), Massie et al.
Grover et al., High Availability Path Design in Ring-Based Optical Networks, 1999, IEEE, vol. 7, No. 4, pp. 558-574.*
IE TF draft entitled, “Carrying Label Information in BGP-4”, Rekhter, et al., 7 pages (Jan. 2000).
IE TF draft entitled, “LDP Specification”, Andersson, et al., 94 pages (Apr. 2000).
IE TF draft entitled “A Framework for Multiprotocol Label Switching”, Callon, et al., 63 pages (Sep. 1999).
IE TF draft entitled Transport of Layer 2 Frames Over MPLS, Martini, et al., 8 pages (Dec. 1999).
IE TF draft entitled “MPLS Label Stack Encoding”, Rosen, et al., 17 pages (Sep. 1999).
MMC Data Sheet entitled “GMAC-B3 Gigabit Media Access Controller”, 50 pages (1999).
MMC Data Sheet entitled “EPIF 4-L3 Reference Manual”, 64 pages (Oct. 1998).
PMC Sierra Inc. Data Sheet entitled “PM-4351 Comet Combined E1/T1 Transceiver/Framer”, Issue 8, 66 pages (Jul. 1999).
Bannai Vinay K.
Barry Charles F.
Choi In-whan
Fan Jason C.
Kalman Robert F.
Lim Krisna
Luminous Networks, Inc.
Sawyer Law Group LLP
LandOfFree
Architecture for transport of multiple services in... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Architecture for transport of multiple services in..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Architecture for transport of multiple services in... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3166542