Electrical computers and digital processing systems: multicomput – Computer-to-computer protocol implementing – Computer-to-computer data transfer regulating
Reexamination Certificate
1998-08-27
2002-06-11
Burgess, Glenton B. (Department: 2153)
Electrical computers and digital processing systems: multicomput
Computer-to-computer protocol implementing
Computer-to-computer data transfer regulating
C709S232000, C709S233000, C709S234000, C709S238000
Reexamination Certificate
active
06405257
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to Internet Protocol (IP) networks, and more particularly, to methods and systems for congestion avoidance in IP backbone networks.
2. Background of the Art
A communications network generally includes a plurality of nodes interconnected by transmission links (or “links”), which are defined by a specific bandwidth (or capacity). In an IP network, a source node transmits data in the form of packets to a destination node through a route, which may traverse zero or more nodes (i.e., switches, routers, bridges, etc.) between the source node and the destination node. The packets may include any type of digitized information, including audio, computer data, video, multimedia, etc. There are several protocols that IP networks, for example the Internet, generally use for transporting data between nodes in the network. One such protocol is the Transmission Control Protocol (TCP).
A TCP/IP network uses a congestion control technique known as Congestion Avoidance Algorithm (CAA). In such a network, nodes use packet losses in the network as congestion indicators and, based on these indicators, adjust their rate of packet transmission. An end node maintains a congestion window based on the congestion indicators that it receives from the network. The node maintains its congestion window size below a maximum segment size per round trip time, and reduces the congestion window size by half each time the source node receives a congestion indicator.
However, random packet losses, which generally have relatively short durations, can cause significant throughput degradation for network services operating in a high bandwidth-delay product regime. Furthermore, in wide area networks, because of long propagation delays, a node cannot adequately track congestions occurring at, for example, a destination node. As a result, the node may significantly decrease its congestion window size, even though the congestion at the destination node no longer exists, resulting in an inefficient allocation and use of network capacity.
Thus, it is desirable to have a method and system for addressing the above and other disadvantages of the existing congestion control avoidance methods for bursty traffic in IP networks.
SUMMARY OF THE INVENTION
Methods and systems consistent with the present invention control congestion in an IP network by preallocating, for a predetermined interval of time, a set of burst access parameters to a set of predetermined routes between a set of source node and destination node pairs in the IP network, and controlling, at the burst level, the traffic at each source node based on the preallocated set of burst access parameters and without communicating with other nodes in the IP network. Specifically, the network controller determines, for a predetermined interval of time, a set of routes between each source-destination node pair in the network, preallocates a set of maximum permitted rates to the preallocated set of routes, respectively, and preassigns a set of burst access thresholds to a set of service classes, respectively, in the predetermined routes, where the pre-assigned burst access thresholds are less than or equal to the corresponding preallocated maximum permitted rates.
Each source node in the network then controls, at the burst level, the external traffic directed to the source node based on the preallocated set of maximum permitted rates and the preassigned set of burst access thresholds and without communicating with other nodes in the network. Specifically, each source node detects the beginning of a burst in the traffic that is accessing that source node. From the predetermined set of routes between the source-destination node pair and based on the peak packet rate, service class, and quality of service (QoS) requirements of the detected burst, the source node identifies a predetermined route for the burst. The source node then identifies the burst access threshold that corresponds to the service class of the burst. The source node admits the detected burst onto the identified predetermined route when a total reserved peak packet rate for all bursts that are in progress in the identified predetermined route plus a peak packet rate of the detected burst is less than the identified burst access threshold. Otherwise, the source node prevents the detected burst from accessing the network.
Methods and systems consistent with the present invention provide a facility for engineering a set of burst access parameters, subject to network capacity, burst rejection quality of service requirements, and conservation flow constraints, such that a total weighted rate of burst rejections in an IP network is minimized. Specifically, the network computes a maximum permitted rate for each predetermined route in the network. Based on the computed maximum permitted rates, the network computes the estimated loads associated with each service class in each predetermined route. Furthermore, based on the computed set of maximum permitted rates and the computed set of estimated loads, the network computes a set of burst access thresholds.
REFERENCES:
patent: 5231631 (1993-07-01), Buhrke et al.
patent: 5335222 (1994-08-01), Kamoi et al.
patent: 5377327 (1994-12-01), Jain et al.
patent: 5572678 (1996-11-01), Homma et al.
patent: 5634006 (1997-05-01), Baugher et al.
patent: 5636212 (1997-06-01), Ikeda
patent: 5668951 (1997-09-01), Jain et al.
patent: 5699346 (1997-12-01), Vandervort
patent: 5812526 (1998-09-01), Chang et al.
Alexander Gersht and Alexander Shulman, “Optimal Routing in Circuit Switched Communication Networks,” IEEE Transactions on Communications, vol. 37, No. 11, Nov. 1989.
Alexander Gersht and Kyoo Jeong Lee, “Virtual-Circuit Load Control in fast Packet-Switched Broadband Networks ,” IEEE, CH2535-3/88/000-0214, 1988.
Gersht Alexander
Pathak Girish
Shulman Alexander
Burgess Glenton B.
Le Hien
Suchyta Leonard Charles
Verizon Laboratories Inc.
Weixel James K.
LandOfFree
Method and system for burst congestion control in an... 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 and system for burst congestion control in an..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and system for burst congestion control in an... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2905599