Multiplex communications – Data flow congestion prevention or control – Control of data admission to the network
Reexamination Certificate
2000-04-11
2004-04-20
Nguyen, Chau (Department: 2663)
Multiplex communications
Data flow congestion prevention or control
Control of data admission to the network
C370S236000
Reexamination Certificate
active
06724725
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to integrated circuit technology and to integrated circuits used in data communications technology. More particularly, the present invention relates to data communications technology utilizing packet-type networked signaling systems. Still more particularly, the present invention will find particular applicability in full-duplex Ethernet Local Area Network (LAN) systems conforming to IEEE standard 802.3x.
2. The Background Art
Ethernet is a CSMA/CD protocol. CSMA/CD (Carrier Sense, Multiple Access, Collision Detect) means that: (1) all stations on the network share the network and have equal access to the same network media (“Multiple Access”); (2) since the network is shared, only one station can transmit on the network at a time and as a consequence, every station has to listen to the network and make sure that no other station is transmitting before that station can transmit (“Carrier Sense”); and (3) in the event that two stations on the network do transmit simultaneously (a condition termed a “collision”), each station on the network must sense the collision and retransmit data later when the network is quiet (“Collision Detect”).
The Institute of Electrical and Electronic Engineers (IEEE) has introduced a series of standards referred to as the 802.3 protocol for implementing Ethernet networks. All Ethernet networks are supposed to conform to this set of standards in order to maintain a base level of interoperability. A portion of this specification is presently entitled “IEEE P802.3x/D3.2 Supplement to Carrier Sense Multiple Access with Collision Detection (CSMA/CD) Access Method & Physical Layer Specifications: Specification for 802.3 Full Duplex Operation” and another portion is presently entitled “IEEE Draft P802.3z/D3.1 Supplement to Carrier Sense Multiple Access with Collision Detection (CSMA/CD) Access Method & Physical Layer Specifications: Media Access Control (MAC) Parameters, Physical Layer, Repeater and Management Parameters for 1000 Mb/s Operation” dated Jul. 18, 1997. These are both hereby incorporated by reference as if set forth fully herein.
The IEEE 802.3 protocol defines the various aspects of a standard Ethernet LAN system.
FIG. 1
is a diagram showing the relationships of the various components of the Physical Layer and Data Link Layer in accordance with the ISO/IEC 8802-3 Open Systems Interconnection (OSI) Reference Model and the IEEE 802.3 CSMA/CD LAN Model. The diagram shows four different implementations: the 1 Mb/s-10 Mb/s version at the left, the 10 Mb/s version, the 100 Mb/s version and the 1000 Mb/s (1 Gb/s) version at the right. In the diagram, from the bottom to the top, the Physical Layer
8
comprises the following: The “MEDIUM”
10
refers to the physical medium over which data is transmitted, i.e., twisted pairs of wires, fiber optic waveguide, coaxial cable, etc. The “MDI”
12
is a Medium Dependent Interface—an interface between the specific medium used (MEDIUM)
10
and the Physical Layer Device (PHY)
14
. The PHY may include a Physical Medium Attachment (PMA)
16
, a Physical Coding Sublayer (PCS)
18
, and a Gigabit Media Independent Interface (GMII)
20
along with a Reconciliation function
22
which reconciles signals between the MAC
26
and the GMII
20
.
At the Data Link Layer
24
, above the Physical Layer
8
, are the Media Access Control (MAC)
26
which controls access to the MEDIUM
10
as through a collision detection system, the optional MAC Control
28
which is responsible for sending and receiving MAC Control Frames for handling flow control; and the Logical Link Control (LLC)
30
which is a software layer controlling the hardware layers below it.
According to this model, the layers interact by way of well defined interfaces, providing services as specified in the IEEE protocol. In general, the interface requirements are shown in FIG.
2
and are as follows:
(1) The interface between the MAC sublayer
26
and its client
31
includes facilities for transmitting and receiving frames, and provides per-operation status information for use by higher-layer error recovery procedures.
(2) The interface between the MAC sublayer
26
and the Physical layer
24
(
FIG. 1
) includes signals for framing (e.g., receive data valid, transmit indication) and contention resolution (e.g., collision detect), facilities for passing a pair of serial bit streams (i.e., transmit, receive) between the two layers.
Communication over an 802.3 LAN occurs between PHY
14
entities. Full-duplex operation allows simultaneous communication between a pair of stations using point-to-point media 10 (dedicated channel). Full-duplex operation does not require that transmitters defer, nor that they monitor or react to collisions, as there is no contention for a shared medium in this mode. Full-duplex is thus available when all of the following are true:
(1) The physical medium is capable of supporting simultaneous transmission and reception without interference;
(2) There are exactly two stations on the LAN. This allows the physical medium to be treated as a full-duplex point-to-point link between the stations. Since there is no contention for use of a shared medium, the multiple access (i.e., CSMA/CD) algorithms are unnecessary; and
(3) Both stations on the LAN are capable of, and have been configured to use, full-duplex operation.
The most common configuration envisioned for full-duplex operation consists of a central bridge (also known as a switch) with a dedicated LAN connecting each bridge port to a single device.
Early Ethernet was based on a bus topology. In a bus topology all stations are connected to one data bus and all stations use the same data bus for transmitting data and receiving data. The first IEEE standardized “bus” media was actually a coaxial cable. This bus topology was well suited for the CSMA/CD protocol. Note that bus topologies (and CSMA/CD) only support half-duplex operation; that is, a station can only transmit data or it can receive data, but it cannot simultaneously transmit and receive data at the same time.
Later, users wanted to migrate away from relatively expensive coaxial cable as a media connection between stations and use common (and less expensive) telephone wire (also known as twisted pair cable) instead. The use of twisted pair cable for Ethernet networks was then standardized by the IEEE and is now commonly designated as “10BaseT”. 10BaseT was different from its coaxial predecessor in that a 10BaseT network uses a “star topology”. A star topology is a network configuration in which each station on the network is connected to a central repeater (also referred to as a “hub”), and there is a separate twisted pair cable for the transmit and receive directions between each station and the repeater. The repeater performed the CSMA/CD functions of repeating a signal from a station to all other stations on the network. This allowed seamless operation with the older bus topology, coaxial based Ethernet.
As time went on, Ethernet became widely deployed and 10BaseT became the medium of choice for networked computers. The success of 10BaseT caused a problem in that networks with lots of users became congested resulting in slow response. Since repeaters basically repeat a transmit packet from one station to all the rest of the stations on the network, a single active station could conceivably dominate the entire bandwidth of the network and other stations could experience significant delays in getting access to the network. To solve this problem, Ethernet switches were invented. Ethernet switches route packets from one station directly to those stations that are supposed to receive the packet. The other stations tied to the switch do not receive the packet and thus can still pass data through the network (unlike a repeater which ties up all stations while one station is transmitting). Hence, the rest of the users are free to access the rest of the network and are not impacted by a few high bandwidth users as much as they would be in repea
Alderrou Donald W.
Dreyer Stephen F.
West Eric T.
LSI Logic Corporation
Ng Christine
Nguyen Chau
Thelen Reid & Priest LLP
LandOfFree
Automatic LAN flow control mechanisms does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Automatic LAN flow control mechanisms, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Automatic LAN flow control mechanisms will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3260894