Multiplex communications – Data flow congestion prevention or control – Control of data admission to the network
Reexamination Certificate
1998-05-12
2002-02-05
Chin, Wellington (Department: 2664)
Multiplex communications
Data flow congestion prevention or control
Control of data admission to the network
C370S395200, C370S395210
Reexamination Certificate
active
06345038
ABSTRACT:
DEFINITION OF TERMS
“Data communication” refers to transmission of digital information. Such information includes, but is not limited to, representations of pictures, sounds, typographic symbols, characters, and numbers. It also includes representations of combinations of these functions; for example, data representing televised sound, video, and closed captioned text. Such representations may have various formats; e.g. compressed and uncompressed formats.
“Access services” refers to services for handling data communications originated by users of a data communication network. A user may be either a person or automated apparatus operating at a station interfacing to the network. Automated apparatus, in the context just used, includes but is not limited to a communication-adapted personal computer or other communication device which can operate either attended or unattended to request access services for a respective station. Access services usually are provided in response to “access request” signals originated by network users.
A “station” in a data communication network is a facility for providing data communication access services. Network routers and administrative control centers are representative of stations in this context.
A network “route” is a data transmission path, through an arbitrary number of stations, between network end users sending and receiving a set of data.
“Peak capacity” refers to the maximum rate of data throughput that a station can sustain. Usually, it is a predetermined function of the aggregate bandwidth available at the station's network interfaces, and the station's internal capacity for storing and handling data.
A station is in a state of “congestion” when its rate of data throughput exceeds a predetermined “limit” near the station's peak capacity.
“Quality of Service” (QoS) pertains generally to bandwidth levels allotted to data sent by individual network users. Such allotments may be implemented by assigning ordered priorities of handling to user's data; e.g. by directing that data initially to different priority stages in a multi-stage queue. A network may contract with its users to provide various levels of QoS ranging from a lowest (“best efforts”) level, wherein data packets of a respective user may encounter delays of arbitrary length at each station in a route, to a highest level that is intended to effect transmission of the data with minimal delays. At stations wherein data is transferred through priority-ordered queues, service applied to data of users subscribing to different QoS levels may be varied by directing the data initially into different priority-ordered stages of the queues.
BACKGROUND OF THE INVENTION
This invention relates to a system and method for improving access to congested networks. Presently, network stations accept requests for new access services until station traffic reaches a defined limit of congestion. Usually, this limit is determined as a function of performance limits of station components and transmission capacities (e.g. bandwidth) of media to which stations interface. When this limit is reached, requests for new or additional access services are rejected until traffic subsides to a level below the congestion limit. The above cross-referenced application by Harrison et al operates in this manner.
This practice, of rejecting new access requests when a station's traffic exceeds its congestion limit, may adversely affect rejected users under circumstances unacceptable to such users; for instance to users seeking to:
execute time-urgent electronic money transfers
execute time-urgent electronic trades in stocks or other securities
complete time-critical work assignments via the Internet (for example while at locations remote from regular business offices)
make time-critical reservations for airline flights, theater production, etc.
obtain immediately needed cash from an automated teller machine
execute time-urgent transactions for a specific business; e.g. a bank or airline reservation firm
The congestion limit may differ for different types of networks. For example, conventional telephone networks might be expected to experience peak traffic demands only at certain critical times; e.g. during certain holidays (Mother's Day, Father's Day, etc.) and at times of major events or disasters such as hurricanes, typhoons, etc. Thus, a properly designed telephone network should be able to handle all traffic at other times and its “limit of congestion” might be based upon traffic expected at such critical times. Similarly, data communication networks like the Internet might be expected to experience unusually heavy traffic loads at critical times. Hence, congestion limits for routers and otherstations in such data networks may be established in contemplation of traffic expected at such times.
SUMMARY OF THE INVENTION
The present invention seeks to alleviate inconveniences to network users caused by the foregoing practice of unconditionally rejecting requests for new or additional access services when traffic at a station receiving such requests is above a defined limit of congestion.
In accordance with this invention, network stations handling traffic exceeding their limit of congestion do not automatically reject requests for new or additional access services. Instead, such new requests are accommodated by degrading access services to selected elements of existing traffic. Users affected by such degrading actions should not be unduly dissatisfied with the resulting service—given to understand that such actions could also benefit them when they otherwise could be blocked from receiving new access services—provided that the periods of degradation are generally brief and that the degradation process is administered on a fair and uniform basis. Financial incentives also could be offered as consideration to users affected by degraded services during times of peak usage.
The degradation process is intended to be “graceful” and have minimal adverse affect on existing traffic. Thus, this process may be applied selectively to elements of traffic receiving a selected class or selected classes of access service. Also, it may be applied on an incremental basis so that degradation of service to an affected element of traffic increases gradually over time.
It is worth noting that degradation processes in accordance with this invention are easily adaptable to function efficiently in stations having different congestion limits, inasmuch as these degradation processes can be applied to selected elements of existing traffic in a manner consistent with the different limits. Furthermore, these degradation processes may also be combined with other procedures to minimize overall impact on the quality of station services. For example, they may be used with (augmented by) other procedures that selectively reject requests for new access services requiring handling of large amounts of data (e.g. multiple megabytes of data), and accept other requests for new access services involving transfers of lesser amounts of data; or even more selectively other procedures that reject only new service requests involving large data transfers at high levels of QoS while accommodating all other requests.
As noted above, degradation of access services to existing data traffic may be applied in a selective manner. Thus, for example, existing traffic subjected to degraded service might consist only of certain types of traffic—e.g. traffic of users entitled to a highest level of QoS or traffic of users currently sending large volumes of data at other than a lowest level of QoS, etc. Also, the degradation may be applied incrementally—e.g. by shifts to incrementally lower priority levels of QoS—resulting in transmittal of affected data at rates that are only incrementally reduced relative to “normal” rates applicable to uncongested conditions.
When volume of existing traffic falls below a predefined “restorations” limit lower than the congestion limit, downgraded services are restored to their “normal” levels (i.e. to levels c
Chin Wellington
Lieber Robert
Tomlin Richard A.
Vong Frank D
LandOfFree
Improving access to congested networks does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Improving access to congested networks, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Improving access to congested networks will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2966368