Electrical computers and digital processing systems: multicomput – Computer-to-computer protocol implementing – Computer-to-computer data transfer regulating
Reexamination Certificate
1996-07-26
2001-03-20
Dinh, Dung C. (Department: 2757)
Electrical computers and digital processing systems: multicomput
Computer-to-computer protocol implementing
Computer-to-computer data transfer regulating
C709S234000, C709S236000, C709S249000, C370S414000
Reexamination Certificate
active
06205486
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates generally to apparatus and method of network interface for data transmission. More particularly, this invention relates to an improved bridge-connector for an inter-network connection provided to dynamically prioritize data frame transfer such that transmission congestion in the network can be minimized.
2. Description of the Prior Art
Data transmission from a high speed network to a low speed network by the use of inter-network connecting devices, e.g., a bridge connector, often encounters the difficulties that transmission congestion may force a network time-out and cause even worse transmission congestion problems. This problem is caused by a conventional data transmission scheme which applies a first-in-first-out (FIFO) prioritization algorithm. This FIFO algorithm is most broadly employed in local area network (LAN), wide area network (WAN), or a connection between FDDI and Ethernet. In the network connections, the conventional FIFO transmission scheme transmits each of the output frames according to the sequence based on the order as these frames are received. When there are greater mismatches between the speeds of the high speed and low speed network, a frame congestion situation is often generated on the side of the low speed network. In the case when a buffer used to store the incoming frames temporarily is filled, the frames that arrive late have to be discarded. A time out for data transmission is often necessary when frames are discarded. Although the problem of a time out is resolved, more data are accumulated during the period when the network is out of service. Greater amount of data are to be transmitted as a result of the time out which often causes an even worse congestion.
FIGS. 1A and 1B
are functional diagrams for illustrating the timing and sequencing of frame transmission over a bridge connector
10
. In
FIG. 1A
, the solid black lines are used to represent transmission of data frames from host computer A in a high speed network
20
to host computer B in a low speed computer network
30
. For the purpose of illustration, a line in the vertical direction represents the time sequence and the width in the horizontal direction represents bandwidth. A normal transmission of frames are shown
FIG. 1A
where three frames A
1
, A
2
, and A
3
are transmitted from the high speed network
20
to the low speed network
30
via the bridge connector
10
. The steeper slopes of A
1
, A
2
, and A
3
on the right-hand side of the bridge connector
10
represent the fact that longer periods of time are required to transmit the frames on the low speed network
30
. The periods T
1
and T
2
represent the time required to send a frame from host computer A to host computer B and to transmit acknowledge signal from computer B to computer A respectively. After the acknowledge signals are received back from the low speed network
30
, more frames are transmitted from the high speed network
20
to the low speed network
30
as shown in FIG.
1
A. This is a typical first-in-first-out (FIFO) transmission process carried out by the bridge connector
10
.
A data transmission congestion is shown in
FIG. 1B
in which the high speed network
20
includes five host computers, computers A, B, C, D, and E, each transmitting data frames, e.g., A
1
, B
1
, C
1
, D
1
, E
1
, A
2
, A
3
, and A
4
, to the low speed network
30
. A frame congestion is generated due to the transmission of five frames on the low speed network
30
. Assume that a communication is carried out between host A in the HSN
20
and a host G in the LSN
30
. Assume also that host-A and host-G use a sliding-windows based on a protocol with “go-back-N” type of error correction for data communication. With the window size of four, a maximum of three outstanding frames between host-A and host-G are allowed. As shown in
FIG. 1B
, the host-A is sending a series of frames A
1
, A
2
, A
3
, etc. to host-G. The first frame Al is sent without delay. Based on the first-in-first-out rule, transmission of A
2
, A
3
, and A
4
has to wait for transmission of B
1
, C
1
, D
1
, and E
1
. As the host-G in the LSN sends the acknowledge signal back to HSN, the time allowed for sending A
2
and A
3
from host-A is over. The host-A then sends frame A
4
instead. Re-transmission of frames A
2
and A
3
from host-A has to be performed later. In addition to the frames transmitted from host computer A, similar situation with frames generated by other computers B, C, D, and E may also occur thus causing further delay and congestion. Re-transmissions of many frames would be required which leads to further degradation of the network performance. When several hosts of the high speed network, e.g., hosts B, C, D, and E, are competing for a low speed data transmission channel, the FIFO priority management scheme applied by a conventional inter-network bridge for handling the timing and sequences of data transmission often causes uncertainties and delays due to the congestion problems. Wastes of precious bandwidth resources are caused by applying the first-in-first-out rule strictly and indiscretionarily.
Therefore, a need still exists in the art of computer network for data transmission to provide a new and improved method which can more effectively utilize the bandwidth of the low speed network to minimize data congestion thus preventing undesirable time-outs so that the difficulties encountered in the art can be resolved.
SUMMARY OF THE PRESENT INVENTION
It is therefore an object of the present invention to provide an improved frame transmission method implemented by the use of an inter-network bridge-connector so that the aforementioned difficulties and limitations in the prior art can be overcome.
Specifically, it is an object of the present invention to provide an improved bridge-connector including a host computer priority queue array wherein the priority of frame transmission is dynamically adjusted adaptive to the transmission state of a network so that the frame congestion can be reduced and the requirement of frame re-transmission can be minimized.
Another object of the present invention is to provide an improved bridge connector including multiple levels of queue arrays where the priority queues are readjusted periodically adaptive to the updated network transmission state so that the random competition of transmission time-windows on a network can be eliminated and priority of frame transmission can be properly arranged according to frame type, byte length and current transmission state of the host computers on the network.
Another object of the present invention is to provide an improved bridge connector wherein higher priority of data transmission are provided to the more frequent hosts in the low speed network during a period so that the data transmission for these hosts in the low speed network can be performed without having to wait for resolution of priority conflict under the rule of FIFO.
Briefly, in a preferred embodiment, the present invention includes a dynamically-adaptive connector-bridge for receiving data frames from a source-network and transmitting the data frame to a destination-network including a plurality of host-computers for receiving the data frames. The connector-bridge includes a data-frame converter for converting the data frames into corresponding destination-network-data-frames suitable for transmission in the destination network. The bridge-connector further includes a dynamic-dispatching means for receiving the destination-network-data-frames for determining a frame-type, a byte-length, and a set of destination-host-computers on the destination-network, and the dynamic-dispatching means employing the frame type, the byte-length, the set of destination-host computers, and a current-data-transmission-state of the destination-network to dynamically prioritize the destination-network-data-frames for dispatching each of the destination-network-data-frames to the set of destination-host-computers.
These and other objects and advant
Huang Gary C.
Wei Edward
Accton Technology Corporation
Dinh Dung C.
Nath & Associates
Novick Harold L.
LandOfFree
Inter-network bridge connector provided for dynamically... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Inter-network bridge connector provided for dynamically..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Inter-network bridge connector provided for dynamically... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2478226