WO2002054680A1 - Dispositif de commande pour un reseau de communication - Google Patents

Dispositif de commande pour un reseau de communication Download PDF

Info

Publication number
WO2002054680A1
WO2002054680A1 PCT/GB2001/005643 GB0105643W WO02054680A1 WO 2002054680 A1 WO2002054680 A1 WO 2002054680A1 GB 0105643 W GB0105643 W GB 0105643W WO 02054680 A1 WO02054680 A1 WO 02054680A1
Authority
WO
WIPO (PCT)
Prior art keywords
network
point
control device
control
point device
Prior art date
Application number
PCT/GB2001/005643
Other languages
English (en)
Inventor
David V. Goadby
Original Assignee
Lanxpress Plc
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Lanxpress Plc filed Critical Lanxpress Plc
Publication of WO2002054680A1 publication Critical patent/WO2002054680A1/fr

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/10—Flow control; Congestion control
    • H04L47/13—Flow control; Congestion control in a LAN segment, e.g. ring or bus
    • H04L47/135—Flow control; Congestion control in a LAN segment, e.g. ring or bus by jamming the transmission media
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L49/00—Packet switching elements
    • H04L49/20—Support for services
    • H04L49/205—Quality of Service based
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L49/00—Packet switching elements
    • H04L49/35—Switches specially adapted for specific applications
    • H04L49/351—Switches specially adapted for specific applications for local area network [LAN], e.g. Ethernet switches

Definitions

  • the field of the present invention relates to a device to be used in a data communications network for controlling the flow of end-user traffic.
  • Ethernet is the most widely used protocol for LAN's (Local Area Networks) and is constantly evolving in order to support modern bandwidth-intensive applications, such as multimedia applications that have video and/or other real-time requirements.
  • LAN's Local Area Networks
  • FDDI Fibre Distributed Data Interface
  • Ethernet as with most other network protocols began as a proprietary protocol, but became standardised with the IEEE 802.3 standard.
  • This standard focuses on the lower layers of the OSI (Open System Interconnection) model, i.e. the MAC (Media Access Control) and PHY (Physical) service interfaces that reside within layers 1 and 2 respectively. Modifications or enhancements to the Ethernet standard are represented using a one or two letter designator.
  • the IEEE 802.3u standard specifies the fast Ethernet standard that was approved in 1995 and is operable at 100Mbps.
  • An even faster Ethernet standard, i.e. the Gigabit (1000Mbps) standard 802.3z has recently been completed.
  • Ethernet is thus a well defined method of networking electronic devices such that they can communicate with each other.
  • access to an Ethernet is uncontrolled due to the fact that it relies on, and expects, simultaneous network access attempts (collisions) as a way of managing itself.
  • Asynchronous data streams require additional processing overhead both to create the data stream and to manage their transmission.
  • the endpoint might manage to successfully transmit a packet of data, which is what is trying to be prevented.
  • the generation of data streams having a preamble requires a special physical device since these signals are part of Layer 2 of the network model.
  • the aim of the present invention is to provide a mechanism for controlling network access in a collision-sensing network alleviating the aforementioned problems, i.e. the present invention has a low processing overhead, is always perfectly timed and can be implemented without the use of non-standard components.
  • a -method of controlling traffic throughput in a collision sensing communications network in which a switching device is connected to a plurality of end-point devices over respective communication channels, the method comprising controlling the traffic throughput over at least one of said communication channels.
  • the controlling step includes selectively isolating at least one end-point device from the network by returning data transmitted by the end-point device back to the end-point device such that the end-point device senses a collision and preventing the network from sensing said collision.
  • the isolation of an end-point device is controlled in the range of 0 to 100% by varying the period of isolation from the network.
  • the communications network is an Ethernet 10/100Base-T network connected in a star topology using unshielded twisted pair cabling and operating in a half-duplex mode.
  • a control device for use in a collision sensing communications network having at least one switching device connected to a plurality of end-point devices over respective communication channels, said control device being adapted to control the traffic throughput in at least one communication channel.
  • the control device comprises: a first physical device connected to transmit and receive channels of said end-point devices; a second physical device connected to transmit and receive channels of said switching device; and a control element for controlling said first and second physical devices.
  • the control element is a processor having an individual network address and may be controlled to set a loopback flag in said first physical device which connects the transmit channel to the receive channel of the end-point device so that an artificial collision is detected by the end-point device.
  • a control device for use in a collision sensing network to selectively isolate an end- point device from the network using a control device located between said end- point device and said network, said control device comprising: a first physical device connected to said end-point device using transmission media over which data can be transferred; a second physical device connected to said network; a control element for controlling said first and second physical devices such that logical switching is performed resulting in data transmitted by the end-point device intended for the network being looped back by the control device to said end- point device.
  • data is looped back at four different locations of the present invention.
  • the first is where data is looped back to the end-point device within said first physical device of the control device.
  • the second is where data is looped back to the end-point device at a location on the transmission media that is external to said first physical device of the control device.
  • the third is where said first physical device has a first interface that connects to the transmission media for receiving data and processes said data such that the processed data may be output on a second interface that is independent of the transmission media and wherein data is looped back to the end-point device at a location on said second interface external to said first physical device.
  • the fourth is where data is read into said control element from said first physical device and is looped back to the end-point device within said control element of the control device.
  • Figure 1 shows a general architecture of a data communications network
  • Figure 2 shows the OSI model
  • Figure 3 shows. the collision sensing mechanism
  • Figure 4a and b show a standard Hub arrangement
  • Figure 5a and b show a switched Hub arrangement
  • Figure 6 shows the general structure of a control device of the present invention
  • Figure 7 shows the location of the PHY
  • Figure 8 shows the transparent mode of operation of the present invention
  • Figure 9a shows the access denied mode of operation of the present invention according to a first embodiment using PHY internal loopback
  • Figure 9b shows the access denied mode of operation of the present invention according to a second embodiment using cable loopback
  • Figure 9c shows the access denied mode of operation of the present invention according to a third embodiment using PHY external digital loopback
  • Figure 9d shows the access denied mode operation of the present invention according to a fourth embodiment using control processor loopback
  • Figure 10 shows the control processor transmit mode of operation of the present invention
  • FIG. 11 shows a further embodiment of the control device of the present invention.
  • Figure 12 shows an example of the control device being implemented throughout a network.
  • Figure 1 shows the general structure of an Ethernet network.
  • Ethernet has various types of topologies.
  • the exemplary topology illustrated in Figure 1 includes a central HUB 4 having a plurality of ports 3 each of which connects to an end-point device 2.
  • the end-point device 2 may be a personal computer (PC) having a NIC (Network Interface Card) allowing an end-user to communicate with the network.
  • PC personal computer
  • NIC Network Interface Card
  • other end-point devices may be provided, such as a network printer having a NIC.
  • OSl Open System Interconnection
  • layers 3-7 are concerned with network details whereas the lower layers 1 and 2 are concerned with the physical details, for example associated with the Ethernet specification.
  • the context of the present invention is concerned with these lower layers. More generally, it can be seen from Figure 2 that layer 2 or the data link layer is associated with a collision detection mechanism (i.e. CSMA/CD) and the establishment of the received data into Ethernet frames.
  • the lowest layer (layer 1) is known as the physical layer and is concerned with the physical aspects of the network, for example the type of network topology used (i.e. star or bus) and/or the type of transmission media being used to transport signals between nodes of a network.
  • the IEEE 802.3 standard provides different technical requirements for the different types of physical media used (i.e. fibre optic, coaxial cable, wireless, etc) where the signals may be optical, electrical or radio frequency signals.
  • Ethernet networks are often named in accordance with their physical characteristics.
  • 10/100Base-T networks use UTP cabling in a star topology operable at 10 Mbps or 100 Mbps respectively.
  • 10Base-2 and 10Base-5 networks both operate at 10Mbps, but use thin coaxial cable and thick coaxial cable for the respective transmission media.
  • 100Base-FX operates at a speed of 100Mbps using fibre optic cable.
  • Figure 3 shows an example of a collision sensing type network, such as an Ethernet network, where a first end-point device B tries to send to a second end- point device A. At the same time a third end-point device C also tries to send a message to device B.
  • the three end-point devices A, B, C are shown in the example as being connected by shared transmission media 30. A collision occurs on the shared media.
  • the sending end-point devices B, C detect the collision, and stop sending. Each retries their respective transmission at a later stage as calculated by the standard exponential back-off algorithm (as specified in the IEEE 802.3 standard).
  • Such an established collision sensing type network of which the Ethernet arrangement described herein is one example, provides a democratic arrangement where all the data packets in circulation are of equal importance, so that they contend with one another for space in the traffic flow.
  • This type of system is well-established and has many advantageous practical applications. The result is that some packets collide and have to be retried at a later time, or some are simply delayed because of their inability to enter the traffic flow at all. However, the end result is that eventually all packets are received and the transfer is complete.
  • Ethernet uses the CSMA/CD collision sensing mechanism and therefore by definition relies on, and expects, collisions as a way of managing itself. Ethernet is therefore not a satisfactory medium for the transmission of real-time traffic.
  • control device of the present invention can be incorporated into a 10/100Base-T network using a half- duplex system.
  • a half-duplex system all stations attached to a shared transmission media are capable of receiving transmissions from all other stations, but the stations cannot receive and transmit data at the same time.
  • full-Duplex systems do not have collisions since all the connections between stations are point to point and therefore data can be transmitted and received simultaneously.
  • the AC104QF package manufactured by Altima CommunicationsTM can be used.
  • FIG. 4a and b An example collision sensing type network is shown in Figures 4a and b and is known as a "hub" or "star" topology.
  • the hub 4 can be thought of as consisting of a plurality of ports, each interfacing with the transmission media that connects to each end-point device.
  • Figure 4a shows a traditional Ethernet hub 4 where a first end-point device A sends a message to a second end-point device B.
  • Figure 4b shows that the Ethernet hub 4 broadcasts the message to all the other end- point devices connected to the hub.
  • considerable bandwidth is used by transmitting each message to end-point devices that are not intended as recipients, i.e. C and D in the example of Figure 4.
  • FIG. 5 shows the same four end-point devices of Figure 4 connected to a switched hub 5.
  • a switched hub uses a high- bandwidth backplane that is typically greater than the sum of the bandwidths supported by each of the network segments fed from the ports of the switch. This allows all ports to use their maximum bandwidth without collisions.
  • Figure 5a shows two end-point devices A and B transmitting data to the switched hub 5 at the same time.
  • Figure 5b shows that the switched hub allows traffic to be switched between any two ports, so that the traffic from end-point device A is switched to its intended recipient C and traffic from end-point device B is switched to its intended recipient D.
  • the use of the switched hub 5 means that each port has its own collision domain and, by being able to control which ports are connected to the sending end-point devices, the segments of the respective receiving end-point devices will not be impacted by unwanted traffic.
  • Switches such as illustrated in Figure 5 are two-fold: i) a network upgrade is costly because traditional hubs will have to be replaced with switched hubs, and ii) switched hubs introduce buffering and switching latency that are especially troublesome for real-time applications.
  • the present invention provides a platform independent solution enabling control of network access and which does not necessitate any change to an existing network infrastructure.
  • the invention enables data with real-time requirements to have priority over other network traffic sessions. This is achieved using a method of controlling access to a collision sensing type network involving inserting a control device between a hub and an end-point device.
  • the control device of the preferred embodiment of the present, invention is a cost effective solution that can be implemented on legacy networks using traditional hubs, while alleviating latency issues.
  • FIG. 6 shows the general structure of a control device 60 for implementing the present invention.
  • the control device 60 consists of two physical devices (PHY) and a processor.
  • a first physical device PHY 64 is connected to an end-point device 2, while a second physical device PHY 62 is connected to the hub 4.
  • a management block 80 is connected to the hub 4.
  • An API block 82 is connected to the control device 60.
  • a control device 60 is preferably placed on each segment fed from a hub and located between the hub 4 and an end-point device 2.
  • the PHY is preferably a single transceiver IC (Integrated Circuit) which is an integral component of the NIC (Network Interface Card) found in any network device or node.
  • the PHY operates at the physical level (i.e. layer 1) and is responsible for controlling the sending and receiving of signals from or onto the chosen transmission media of the network.
  • the PHY manages the physical connection to the network and checks that all the Ethernet signals adhere to the IEEE 802.3 specifications. The operation of such PHY devices is well-known.
  • the PHY integrated circuit to be used for the control device of the present invention must be able to support a so-called "remote loopback" function, for example Altima Communications' AC101 Ethernet transceiver that uses line 3 of register 24 (i.e. Reg. 24.3) for this purpose.
  • the remote function, of the PHY allows incoming data to be equalised and then looped back to the transmit driver.
  • FIG. 7 is a block diagram showing the location of the PHY. It can be seen that the PHY 62, 64 resides between . the transmission media block 74 that connects to the network and the MAC or micro-processor block 70. Moreover, the interface 72 connecting the MAC 70 and the PHY 62, 64 contains two important signals, which are the Carrier Sense signal (CRS) and a Collision Detect Signal (COL). Therefore, the NIC uses the PHY to monitor the network status and return the status of the network using these signals. Thus, the MAC or microprocessor unit may check these signals before attempting to transmit traffic to the network.
  • CRS Carrier Sense signal
  • COL Collision Detect Signal
  • Loopback is defined as the return of data transmitted from an Ethernet end-point device to that same device. The data is not altered in any way, just returned. Loopback should only be set when the network is idle, i.e. not transmitting any data.
  • a Carrier Sense signal of the PHY is asserted when the network connection to the end-points device with which it is associated is not idle (i.e. busy) and is used to determine whether loopback can be enabled.
  • FIG. 6 shows the loopback function of an end-point device PHY operating in the control device 60 of the present invention where the transmitted data from the end-point device (indicated by arrow A) is returned to the end-point device (indicated by arrow B) if the loopback register of end-point device PHY has been set.
  • the setting and resetting of the loopback functionality of the end-point device PHY 64 is carried out by a control program executed on the control processor 66 connected to the two physical devices. This can either be done locally, i.e. on the control device 60 itself or alternatively at a remote location.
  • the control processor 66 is shown to contain a MAC (Media Access Control) address 68 and therefore can be configured to be an addressable node of the network thereby facilitating control from a remote location.
  • MAC Media Access Control
  • control device 60 For a better understanding of the present invention, the three modes of operation of the control device 60 are described hereinbelow with references to Figures 8, 9 and 10.
  • the control device 60 is shown to have two back-to- back PHY's 62, 64 being coupled using a block of functionality referred to herein as the control logic 63.
  • the control logic 63 shown in Figures 8, 9 and 10 is a collection of logic gates that steer data depending on the mode of operation of the control processor 66.
  • Figure 8 shows that the control device 60 has no effect on the data traffic to or from the end-point device 2.
  • the control logic 63 is such that the control processor 66 is able to receive (listen to) information intended for the end-point device 2 thereby allowing the control device 60 to monitor network traffic.
  • control processor 66 has its own MAC 68 and is therefore an individually addressable device of the network in its own right.
  • the control processor 66 shares the receive path with the actual end-point device 2 so that it may respond to any network traffic addressed to it. If the control processor is addressed and needs to respond, then once the current network session is completed the control processor 66 will change the control logic 63 to place the control device 60 into a third mode of operation ( "control processor transmit mode"), which is described in detail hereinbelow.
  • FIGS 9a to 9d illustrate the end-point access denied mode of operation.
  • the control device 60 is able to carry out this mode of operation using loopback.
  • Loopback can be carried out at four main locations in the datapath where the effect of loopback, i.e. to create artificial collision traffic as seen by the end-point device 2, is the same no matter where it is applied.
  • the four locations of loopback are referred to herein as "PHY internal loopback”, “cable loopback”, “PHY external digital loopback” and “control processor loopback” respectively.
  • Figure 9a shows PHY internal loopback, which occurs within the end point PHY 64.
  • Some chips for example Altima's AC101
  • Altima's AC101 have such loopback facilities that can be enabled under software control. From the end-point device PHY's 64 perspective, data is received from an end-point device 2, and if the remote loopback flag on register 24.3 of Altima's AC101 chip is set then the data is returned to the end-point device 2, as illustrated by the loopback arrow 61.
  • the second loopback location (cable loopback) is shown in Figure 9b as anywhere on the transmission media or physical layer connecting the end-point device 2 to the control device 60.
  • the cable connections are crossed over, i.e. transmit (Tx) to receive (Rx) and vice versa. This is illustrated by the loopback arrow 67.
  • a simple implementation of this might be done using relay switches, although high speed semiconductor switches which handle high speed data without introducing any delay would be the preferred solution.
  • a cross-over switch 65 is illustrated in Figure 9b.
  • the third loopback location (PHY external digital loopback) is shown by loopback arrow 69 in Figure 9c at the Mil (Media Independent Interface) of the PHY.
  • This interface resides at the back end of the PHY and is independent of the type of transmission media used by the network. That is, data is transmitted and received over the physical cables in analogue form, but the Mil deals with the digital data which is received after decoding and transmitted before encoding. Therefore, to achieve loopback at the Mil location, the Mil received data is connected to the transmit data which creates a return data path.
  • the fourth loopback location (control processor loopback) is shown by loopback arrow 92 within the control processor 66.
  • data is read from the end- . point device PHY 64 into the control processor and then retransmitted back to the same PHY (i.e. the end-point device PHY 64) under program control.
  • Figure 9a described hereinabove illustrates the case when the loopback register of the end-point device PHY 64 is set.
  • the action of setting the remote loopback register in this scenario has two major effects: a) The end-point device 2 senses a collision and therefore stops transmitting information. At a later stage it retries the transmission. If it is assumed that end-point device PHY 64 is maintained in loopback mode, a collision is again sensed and the end-point device will back off once again.
  • the artificial collision triggers a mechanism that stops the sending device from transmitting for a variable amount of time dependent on the exponential back-off algorithm defined in the IEEE 802.3 standard. After a few attempts at transmission the end-point device will not transmit at all and the current network session will be abandoned altogether.
  • the network side of the control device 60 sees no traffic emanating from the end-point device 2 and therefore the bandwidth available to the network may be used for other network traffic.
  • the bandwidth available to the network may be used for other network traffic.
  • the control processor 66 can still respond to network traffic since it is a device in it's own right. More generally, the control processor 66 is still able to receive network traffic and can respond if required by changing to the third mode.
  • Figure 9b described hereinabove illustrates the case when transmitted data is connected to receive data at the cable level thereby creating the required artificial traffic for the end-point device 2.
  • the end-point device PHY 64 does not see any data when this loopback method is used, which also implies that the network PHY 62 does not see any data either.
  • the control processor 66 is still able to receive network data and can respond if required by changing to the third mode.
  • Figure 9c described hereinabove illustrates the case when data received by the end-point device PHY 64 is immediately transmitted back to the end-point device 2 to force artificial collision traffic.
  • the end-point device PHY 64 does not pass the received data to the network PHY 62, so no traffic is transmitted to the network.
  • the control processor 66 is still able to receive network data and can respond if required by changing to the third mode.
  • Figure 9d described hereinabove illustrates the case when data is read into the control processor 66 from the end-point device PHY 64 and is then retransmitted back to the same PHY 64 under program control of the control processor 66.
  • the control processor 66 does not pass data to the network PHY 62, so no traffic is transmitted to the network.
  • the control processor does incur some overhead while retransmitting the read data, but is still able to receive network data and can respond if required by changing into the third mode as described below.
  • Figure 10 illustrates the control processor transmit mode of operation.
  • the control logic 63 is such that the control processor 66 is able to transmit and receive. Therefore, in this mode of operation the control processor 66 can initiate, or respond to, network traffic. However, since the network connection is shared with the end-point device this potentially leads to contention. Therefore, when the control processor 66 requires network access, the control logic 63 is arranged to connect the network PHY 62 to the control processor 66 only. However, since data may need to be sent to the end-point device, the control logic also needs to provide a path to connect the transmitted data from the controi processor to the end-point device PHY 64.
  • test mode of operation might be invoked whereby the loopback function associated with the two PHY's 62, 64 of the control device 60 could be used in a conventional manner. That is, loopback as a diagnostic function could be used when the whole of the network, or at least the collision domain, is being tested. The use of this facility would preferably be only for the testing of a network during installation or engineering changes. Mode changes are carried out by the control processor 66 and have to be done carefully since data loss could occur if changes were made without waiting for any existing network sessions to be completed. The network status is read from the PHY's (using the CRS and COL signals) where both the network PHY 62 and end-point device PHY 64 must be checked by the control processor 66 before taking any action.
  • FIG. 6 also shows a manager element 80, which for example could be a terminal running a management application connected to the network through the hub 4.
  • the manager 80 could monitor various data traffic parameters passing through each control device 60, for example the flow or bandwidth used by an end-point device 2. So for example, the manager may have predetermined traffic profiles which could be achieved by isolating or controlling the amount of traffic generated by selected end-point devices 2 by remotely changing the relevant control devices into their end point access denied modes.
  • Controlling the period of time that the loopback function relating to a selected control device is set determines the data throughput generated by the corresponding end-point device. So, each control device 60 can be managed so that the bandwidth consumed by the relevant end-point device 2 is managed from 0 - 100%.
  • the action of changing modes of operation is carried out by a control processor 66 that executes an associated software control program, which is responsible for initiating respective modes based upon certain conditions. As an example these conditions include, but are not limited to:
  • Time triggered events for example at the end of the working day.
  • Figure 6 shows an API (Application Programmer Interface) 82 that can be connected to the control processor 66 to allow potential buyers of the control device 60 to be able to tailor the front-end of the control program software executed on the processor 66 to meet individual market requirements. Therefore by using an option socket and a software API, additional features can be used.
  • API Application Programmer Interface
  • the basic mother board assembly of the control device 60 is intended to provide the basic service of priority channelling by adjusting the traffic flow of selected end-point devices between 0-100%.
  • Other facilities that are included in the basic unit are:
  • MIB Management Information Base
  • VoIP Voice over Internet Protocol
  • - Streaming audio interface i.e. an open standard developed for wireless (RF) connectivity
  • USB Universal Serial Bus
  • - IEEE 1284 interface i.e. for supporting an 8-bit high-speed bi-directional port.
  • Two additional Ethernet devices can be attached including a fibre channel.
  • Smartchip security access which are enhanced smartcard devices that include special sensors. For example, fingerprint, retinal image and signature.
  • Figure 11 shows a further embodiment of the present invention, where the control device 60' no longer has a programmable processor 66, but instead uses . dedicated hardware 92 to couple the end-point device PHY 64 and the network PHY 62.
  • the dedicated hardware may take many forms, such examples might be: a specialised IC (Integrated Circuit), electronic relay switching logic or even a dedicated control processor where the software control program is fixed.
  • switching logic within the control device 60' can be customised to encompass the whole PHY-PHY link area and could even have embedded chips in it as indicated by the integrated PHY's of Figure 11.
  • the end result for this embodiment is the same as the previously described embodiments of the access denied (loopback) mode of operation, which is to 'echo' all transmitted data from the end-point device back to itself.
  • Figure 12 shows the control device 60 being implemented in a network.
  • these devices 60 would be positioned on all the segments of a particular network thereby allowing any end-point device attached to the network to be isolated.
  • hubs and other network legacy devices may be connected to the hubs.
  • the hub 4' is cascaded from the central hub 4.
  • Other network nodes may include bridges, routers and repeaters.
  • the control device can similarly be placed in these networks as required. If a control device 60 is placed, as shown in Figure 9 between the main hub 4 and cascaded hub 4' and the control device is set to loopback, then all the devices downstream from the main hub and attached to the cascaded hub will be isolated. Thus, a single control device will completely isolate the cascaded network.
  • control devices 60 should be positioned in all the tributary segments emanating from the cascaded hub 4'. Therefore, when all the network devices are fitted with the present invention, the individual data paths connecting each network device can be managed to provide pre-set bandwidths so that high speed data will get priority over low speed data.
  • the end-point devices may be isolated from a network, by creating artificial collision traffic as seen by selected devices. This enables the bandwidth of the network to be managed thereby allowing the transmission of data to be regulated to specific requirements. So, although network traffic can be a mi of slow speed data (for example, a printer) and high speed data (for example, video or audio), the present invention allows a network to be managed such that higher priority traffic is delivered more reliably than on an unmanaged system.
  • the removal of the remote loopback function from the end-point device PHY 64 restores the end-point device 2 to normal operation with no side effects or loss of data.
  • the list of applications and additional features that have been described is not an exhaustive list where the idea of the API has been specifically introduced to allow programmers the flexibility of 'tweaking' the control device in order to support their requirements or satisfy any emerging technologies.
  • the transceiver chip i.e. the AC104QF manufactured by Altima CommunicationsTM
  • the loopback function is performed external to the PHY, for example using either the second, third or fourth embodiments of the access denied mode of operation of the present invention as described hereinabove (see Figures 9b, 9c and 9d respectively).
  • the control devices are shown located between the hub and the end-point devices, this is a logical positioning and in practice the functionality of the control devices may be incorporated within the hub itself.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Small-Scale Networks (AREA)

Abstract

L'invention concerne un dispositif de commande situé entre les appareils terminaux d'un réseau de communication et un noeud concentrateur central permettant de commander la vitesse du cheminement de données entre 0 et 100 %. En commandant la quantité de trafic vers et/ou de chaque appareil terminal, le réseau de communication peut créer la largeur de bande pour d'autres applications dont la priorité peut primer, par exemple, des données vidéo. Selon un mode de réalisation favori, le dispositif de commande est composé de deux appareils physiques dos à dos (PHY) commandés par un processeur possédant une adresse de réseau. Les premières interfaces PHY avec le réseau et les autres interfaces PHY avec l'appareil terminal. L'appareil PHY terminal peut être réglé en mode bouclage, de manière à ce qu'un trafic de collision artificiel soit vu par l'appareil terminal qui interrompra alors la production de trafic.
PCT/GB2001/005643 2000-12-28 2001-12-19 Dispositif de commande pour un reseau de communication WO2002054680A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
GB0031751.1 2000-12-28
GB0031751A GB2374496A (en) 2000-12-28 2000-12-28 Control device for a collision-sensing communication network

Publications (1)

Publication Number Publication Date
WO2002054680A1 true WO2002054680A1 (fr) 2002-07-11

Family

ID=9905963

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/GB2001/005643 WO2002054680A1 (fr) 2000-12-28 2001-12-19 Dispositif de commande pour un reseau de communication

Country Status (2)

Country Link
GB (1) GB2374496A (fr)
WO (1) WO2002054680A1 (fr)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP3310067B1 (fr) * 2016-10-14 2020-02-26 Alcatel Lucent Système de gestion de chemin de données, unité de commande et unité de commutation

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO1997043843A1 (fr) * 1996-05-15 1997-11-20 Medialink Technologies Corporation Procede et appareil permettant de reguler le flux de donnees dans un reseau ethernet
WO1998011695A1 (fr) * 1996-09-16 1998-03-19 Advanced Micro Devices, Inc. Interface independante des supports pour interfaces de couches physiques ethernet rapide 802.3u (100 base t)
EP0865179A1 (fr) * 1997-03-13 1998-09-16 Siemens Business Communication Systems, Inc. Procédé et système de gestion des collisions dans un réseau basé sur un concentrateur
WO2000033510A2 (fr) * 1998-11-30 2000-06-08 International Business Machines Corporation Systeme de traitement de donnees, et procede permettant de d'invalider a distance l'activite d'un reseau dans un systeme informatique de client
WO2000059176A2 (fr) * 1999-03-31 2000-10-05 Broadcom Corporation Appareil pour communication ethernet phy/mac

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE3462300D1 (en) * 1983-05-06 1987-03-05 Toshiba Kk Packet communication system
US5802054A (en) * 1996-08-15 1998-09-01 3Com Corporation Atomic network switch with integrated circuit switch nodes
US6178483B1 (en) * 1997-02-14 2001-01-23 Advanced Micro Devices, Inc. Method and apparatus for prefetching data read by PCI host
US6215789B1 (en) * 1998-06-10 2001-04-10 Merlot Communications Local area network for the transmission and control of audio, video, and computer data

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO1997043843A1 (fr) * 1996-05-15 1997-11-20 Medialink Technologies Corporation Procede et appareil permettant de reguler le flux de donnees dans un reseau ethernet
WO1998011695A1 (fr) * 1996-09-16 1998-03-19 Advanced Micro Devices, Inc. Interface independante des supports pour interfaces de couches physiques ethernet rapide 802.3u (100 base t)
EP0865179A1 (fr) * 1997-03-13 1998-09-16 Siemens Business Communication Systems, Inc. Procédé et système de gestion des collisions dans un réseau basé sur un concentrateur
WO2000033510A2 (fr) * 1998-11-30 2000-06-08 International Business Machines Corporation Systeme de traitement de donnees, et procede permettant de d'invalider a distance l'activite d'un reseau dans un systeme informatique de client
WO2000059176A2 (fr) * 1999-03-31 2000-10-05 Broadcom Corporation Appareil pour communication ethernet phy/mac

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP3310067B1 (fr) * 2016-10-14 2020-02-26 Alcatel Lucent Système de gestion de chemin de données, unité de commande et unité de commutation

Also Published As

Publication number Publication date
GB2374496A (en) 2002-10-16
GB0031751D0 (en) 2001-02-07

Similar Documents

Publication Publication Date Title
US5982778A (en) Arrangement for regulating packet flow rate in shared-medium, point-to-point, and switched networks
US6198722B1 (en) Flow control method for networks
US5754552A (en) Automatic communication protocol detection system and method for network systems
US5121382A (en) Station-to-station full duplex communication in a communications network
US5574722A (en) Protocol independent switch
US7428586B2 (en) System and method for discovering undiscovered nodes using a registry bit in a point-to-multipoint network
US5537099A (en) Receiving port security in a network concentrator
EP0939511A2 (fr) Procédé et système de régulation automatique du débit d'une liaison dans un répéteur de réseau
US6256321B1 (en) Information communication network system, central information communication control device and information communication device used in the system, information sending method, and modulation method
US6690650B1 (en) Arrangement in a network repeater for monitoring link integrity by monitoring symbol errors across multiple detection intervals
US20030177216A1 (en) Isolation technique for networks
US6484213B1 (en) Adapting networking device for enhancing performance of a hybrid networking system
KR20020069562A (ko) 한 쌍의 신호선을 통한 고속 데이터 통신이 가능한 이더넷시스템
US8184677B1 (en) Arrangement in a network repeater for monitoring link integrity and automatically down shifting link speed
Cisco Out-of-Band Management
Cisco Out-of-Band Management
Cisco FDDI Full-Duplex Port Adpaters Feature Module
Cisco FDDI Full-Duplex Single-Mode and Multimode Port Adapters
Cisco FDDI Full-Duplex Single-Mode and Multimode Port Adapters
WO2002054680A1 (fr) Dispositif de commande pour un reseau de communication
Cisco PA-4R-FDX Token Ring Full-Duplex Port Adapter Feature Module
Cisco PA-4R-FDX Token Ring Full-Duplex Port Adapter
Cisco PA-4R-FDX Token Ring Full-Duplex Port Adapter
Cisco PA-4R-FDX Token Ring Full-Duplex Port Adapter
Cisco PA-4R-FDX Token Ring Full-Duplex Port Adapter

Legal Events

Date Code Title Description
AK Designated states

Kind code of ref document: A1

Designated state(s): AE AG AL AM AT AU AZ BA BB BG BR BY BZ CA CH CN CO CR CU CZ DE DK DM DZ EE ES FI GB GD GE GH GM HR HU ID IL IN IS JP KE KG KP KR KZ LC LK LR LS LT LU LV MA MD MG MK MN MW MX MZ NO NZ PL PT RO RU SD SE SG SI SK SL TJ TM TR TT TZ UA UG US UZ VN YU ZA ZW

AL Designated countries for regional patents

Kind code of ref document: A1

Designated state(s): AM AZ BY KG KZ MD RU TJ TM AT BE CH CY DE DK ES FI FR GB GR IE IT LU MC NL PT SE TR BF BJ CF CG CI CM GA GN GQ GW ML MR NE SN TD TG

121 Ep: the epo has been informed by wipo that ep was designated in this application
REG Reference to national code

Ref country code: DE

Ref legal event code: 8642

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: "NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 69(1) EPC (F1205A DATED 04.12.03)

122 Ep: pct application non-entry in european phase
NENP Non-entry into the national phase

Ref country code: JP

WWW Wipo information: withdrawn in national office

Country of ref document: JP