Network traffic management

Details Network traffic management Network traffic management

1993-05-10

1994-12-06

Olms, Douglas W.

Multiplex communications

Wide area network

Packet switching

370 60, 370 79, 370 941, 371 32, H04J 314

Patent

active

053717317

DESCRIPTION:

BRIEF SUMMARY
BACKGROUND OF THE INVENTION

1. Field of the Invention
This invention relates to traffic management in a relay-based communications network.
2. Related Art
By "relay-based communications network" is meant any network using a relaying technique for transmitting, for example, blocks or frames of data, such as in nodal networks for packet switching systems. Examples of these are frame mode bearer service (FMBS), as defined in the International Consultative Committee for Telephone and Telegraph (CCITT) publication currently called "The Blue Book" which is a set of recommendations for telecommunication standards, Broadband-ISDN, cell relay and bridged local area networks.
In an FMBS the information is transmitted in frames comprising an address field A, a control field C, an information field I and an error check fiend CRC. This is illustrated in FIG. 1.
Congestion management policy for FMBS has been proposed by the CCITT as recommendation I. 3XX in the latest version of the above publication.
The header of a frame for use in FMBS has a flag in it called the "discard eligibility indicator" (DEI). If a node is congested and some frames must be discarded to ease the problem, the frames with the DEI flag set are chosen for discard in preference to frames not having this flag set. This flag can be set by the source terminal (i.e. the originator of the frames) or the ingress node (i.e. the first node on the virtual circuit through the network). In theory, for any call the proportion of frames with the DEI set can vary between 0 and 100%. The proportion is determined by the quality of service requested by the source terminal and confirmed
egotiated by the network when the virtual call is established. Thus, in the event of congestion, a decision as to which frames to discard is based solely on whether the DEI flag is set. No regard is paid to such things as the maximum delay negotiated by the call at the time of set up.
Correct reception of a frame or a group of frames is confirmed by the destination terminal transmitting a positive acknowledgement signal to the source terminal. Faulty transmission in such a system may include, for example, corrupted data due to noise and/or lack of reception of the data at the correct address. Corrupted data may give rise to a negative acknowledgement signal (NAK), as opposed to a positive one, transmitted from the destination terminal to the source terminal. At, for example, 2 Mbit/sec transmission rates, due to the delay between detection of a corrupted frame and receipt of the NAK at the source terminal, a number of subsequent frames will have been transmitted between the same source and destination terminals before correction of the error can be initiated. Upon the absence of the expected transmitted frame thus giving rise to the NAK, the destination terminal will discard all subsequently received frames until the frame in error is received correctly. Thus, the source terminal employs a "GO BACK N" protocol on receipt of a N which identifies the last correctly received frame and thus the N frames needing to be retransmitted. Hence the NAK causes retransmision of the N frames following the last correctly received frame. After retransmission, transmission of subsequent frames is resumed.
Because frames that will, in any event, be discarded by the destination terminal are transmitted through the network, congestion is exacerbated by the unnecessary transmission of the redundant frames.
Furthermore, and more importantly to the customer, it is not possible to bill accurately for successfully transmitted frames as it is not possible to discriminate between legitimately received frames and those received and rejected subsequent to transmission of a previously discarded frame. Billing monitoring is normally located at the egress node in a network (i.e. that adjacent the destination terminal in a virtual call route) which simply notes the reception of frames without being able to distinguish between original transmission of frames, which should be written off, and retransmission of those

REFERENCES:
patent: 5042029 (1991-08-01), Hayakawa
patent: 5090011 (1992-02-01), Fukuta et al.
patent: 5245616 (1993-09-01), Olson
PCT International Search Report.
IEEE Network: The Magazine of Computer Communication, vol. 2, No. 1, Jan. 1988, New York US, pp. 72-76; Gerla et al: "Congestion Control in Interconnected Lans".

Affiliated with

Also associated with

Rating

Network traffic management does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Network traffic management, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Network traffic management will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-219780