WO1996018278A1 - Large capacity atm switch - Google Patents
Large capacity atm switch Download PDFInfo
- Publication number
- WO1996018278A1 WO1996018278A1 PCT/CA1995/000184 CA9500184W WO9618278A1 WO 1996018278 A1 WO1996018278 A1 WO 1996018278A1 CA 9500184 W CA9500184 W CA 9500184W WO 9618278 A1 WO9618278 A1 WO 9618278A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- input
- cells
- port
- cell
- ports
- Prior art date
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/54—Store-and-forward switching systems
- H04L12/56—Packet switching systems
- H04L12/5601—Transfer mode dependent, e.g. ATM
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L49/00—Packet switching elements
- H04L49/10—Packet switching elements characterised by the switching fabric construction
- H04L49/104—Asynchronous transfer mode [ATM] switching fabrics
- H04L49/105—ATM switching elements
- H04L49/106—ATM switching elements using space switching, e.g. crossbar or matrix
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L49/00—Packet switching elements
- H04L49/25—Routing or path finding in a switch fabric
- H04L49/253—Routing or path finding in a switch fabric using establishment or release of connections between ports
- H04L49/255—Control mechanisms for ATM switching fabrics
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L49/00—Packet switching elements
- H04L49/30—Peripheral units, e.g. input or output ports
- H04L49/3081—ATM peripheral units, e.g. policing, insertion or extraction
- H04L49/309—Header conversion, routing tables or routing tags
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L49/00—Packet switching elements
- H04L49/45—Arrangements for providing or supporting expansion
- H04L49/455—Provisions for supporting expansion in ATM switches
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/54—Store-and-forward switching systems
- H04L12/56—Packet switching systems
- H04L12/5601—Transfer mode dependent, e.g. ATM
- H04L2012/5603—Access techniques
- H04L2012/5604—Medium of transmission, e.g. fibre, cable, radio
- H04L2012/5605—Fibre
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/54—Store-and-forward switching systems
- H04L12/56—Packet switching systems
- H04L12/5601—Transfer mode dependent, e.g. ATM
- H04L2012/5638—Services, e.g. multimedia, GOS, QOS
- H04L2012/5646—Cell characteristics, e.g. loss, delay, jitter, sequence integrity
- H04L2012/5651—Priority, marking, classes
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/54—Store-and-forward switching systems
- H04L12/56—Packet switching systems
- H04L12/5601—Transfer mode dependent, e.g. ATM
- H04L2012/5672—Multiplexing, e.g. coding, scrambling
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/54—Store-and-forward switching systems
- H04L12/56—Packet switching systems
- H04L12/5601—Transfer mode dependent, e.g. ATM
- H04L2012/5678—Traffic aspects, e.g. arbitration, load balancing, smoothing, buffer management
- H04L2012/5681—Buffer or queue management
Definitions
- This invention relates generally to large capacity ATM switches.
- it is directed to ATM switches in which high speed links connect input/ output buffers and a space switch core, and connections are set up through the space switch core whenever trains of cells are to be transmitted between the input and output buffers.
- High capacity fiber transmission already provides the technical means to move very large amounts of data from node to node at reasonable cost.
- Broadband access systems are based on systems such as fiber to the home, fiber to the curb, coaxial cable, or wireless, and will serve to connect subscribers to local access nodes.
- ATM asynchronous transfer mode
- the construction of an efficient access and tandem network will require very large ATM switches with aggregate capacities in the 100's to 1000's of gigabits per second (Gbit/s).
- Average peak hour demand may range from a few 100 kbit/s to 10 Mbit/s or more per subscriber, depending upon the offered service. For example, digital video-on- demand, using MPEG2, could easily generate a network demand of 5 or 10 Mbit/s per household (the bottleneck in this scenario appears to be the video server capacity).
- a timeslot utilization means is provided in an ATM switch for scheduling the earliest possible connection between an input port and output ports.
- a timeslot scheduling unit in an ATM switch prevents the packets from collision in a space division switching unit.
- Each packet buffer unit at each port writes packets sequentially but reads out randomly in the timeslots assigned by the timeslot scheduling unit so that the throughput of the space division switching unit is improved.
- an ATM switch includes ATM line terminating units and a self- routing space switch.
- the overall architecture of the switches of the present invention is based loosely on a space switch crosspoint, input and output buffers and substantially high speed links connecting them. None of the above patents is concerned with packet switches in which a train of one or more ATM cells are transmitted by way of high speed links through a space crosspoint in one burst.
- the invention is directed to a high capacity ATM switch for switching data between input ports and output ports in a variable length train of digital cells, each input and output port handling one or more channels and each cell comprising a fixed number of bytes and including a channel indication therein.
- the switch includes an input buffer at each input port for separately storing trains of digital cells destined for output ports and for sending a connection request through a control link to switch core means.
- the switch core means makes and breaks high speed data link connections between the input ports and the output ports at desired times and indicates by way of the control link to each input buffer when to send the train of digital cells stored therein.
- the switch also includes an output buffer at each output port which separately stores cells of bytes received from the switch core means and destined to the channel in response to the channel indication of each cell.
- the invention is directed to a method of switching data in ATM mode between input ports and output ports in a variable length train of digital cells, each input and output port handling one or more channels and each cell comprising a fixed number of bytes and including a channel indication therein.
- the method comprises steps of separately storing, at an input buffer of each input port, trains of digital cells destined for output ports and sending a connection request through a control link to a switch core.
- the method further includes steps of the switch core means, in response to the connection request from one or more input buffers, making and breaking high speed data link connections between the input ports and the output ports at desired times and indicating by way of the control link to each input buffer when to send the train of digital cells stored therein.
- the method also includes a step of separately storing, at an output buffer of each output port, cells of bytes received from the switch core means and destined to the channel in response to the channel indication of each cell.
- FIG. 2 is a schematic illustration of an ATM switch using a novel expansion scheme according to the invention
- FIG. 3 shows basic components of the ATM switch according to the invention
- FIGS. 4-7 show various parts of the ATM switch of one embodiment of the invention in more detail
- Figure 8 shows a hardware arrangement of the space switch core of the invention
- Figure 9 shows schematically an optical shuffle according to one embodiment of the invention
- Figure 10 is a schematic illustration of an optical data selector according to one embodiment
- Figure 11 shows the common core control mechanism in block diagram
- Figure 12 in a schematic illustration of a distributed core control mechanism
- FIG 13 shows a further embodiment of the invention in which a WDM (Wavelength Division Multiplexing) technique is used for sending data and control streams through a single fiber optic; and
- WDM Widelength Division Multiplexing
- Figures 14 and 15 are graphs showing delay distribution and burst length distribution respectively. Detailed Description
- an ATM switch comprises a switch core, peripheral units with transmission interfaces, and a software control structure running on embedded and /or external CPUs.
- a large capacity ATM switch is created by expanding this architecture by inserting independent buffered satellite switches between groups of lower speed access peripherals and the switch core.
- Figure 1 illustrates a widely utilized three stage expansion scheme for a large capacity switch.
- a 100 Gbit/s capacity switch can be constructed by using thirty 10 Gbit/s switches in three stages, with 1 Gbit/s links connecting them together. A total of 200 links are required.
- Figure 2 is a schematic illustration of a novel expansion scheme which is used to construct the large capacity ATM switch of the invention.
- Each buffered switch can be considered as a stand-alone switch and a large switch is created by tying together a cluster of smaller switches using very high speed (e.g. 10 Gbit/s) trunks.
- the middle stage of the three stage expansion scheme is replaced by a space switch core with no buffering.
- the space switch core sets up connections but, instead of processing one ATM cell at a time as in the scheme of Figure 1, each outer stage switch transmits a train of more than one ATM cell at a time through the space switch core.
- the space switch core control can be operated at a slower speed. Twenty smaller switches and 20 links are required to construct a switch of the same capacity as the one shown in Figure 1.
- the switch 10 is shown in the unfolded configuration and includes N input buffer modules 12, N output buffer modules 14, and an N port space switch core 16.
- Each input buffer module consists of a cell RAM 18, a multiplexer 20 and a S/P (serial-to-parallel) converter and multiplexer 22.
- a buffer control 24 is also shown.
- each output buffer module there are a cell RAM 26, a multiplexer 28 and P/S (parallel-to- serial) converter and multiplexer 30.
- a space switch core 16 connects the input and output modules under the control of a core controller 32. It should of course be noted that ordinarily an input buffer module and an output buffer module are located at a port and therefore the cell RAM at each buffer module can be made of a single memory with a time divided buffer control circuit serving both functions.
- the switch carries ATM cells transparently from the input buffer modules to the appropriate output buffer modules through the space switch core via the data link 34.
- the input buffer module receives incoming traffic in ATM cells and logically segregates them into FIFOs (pages) according to port destination, and then delivers a train of cells from the selected FIFO (page) to the space switch through the data link. Cells can also be prioritized also for different services if needed. Cell routing information which is inherent in ATM header VCI/VPI fields is translated into switch specific port address information in the input buffer modules and prepended to the standard 53-byte cells. Since there are no cell buffers in the space switch core, transmission into the space switch core from the individual buffer modules is coordinated in such a way that cells do not collide at the space switch core outputs.
- control controller This is achieved by using a common hardware controller circuit (core controller) 32 which receives status information (e.g. a connection request) from the input buffer modules. When requested by input buffer modules, the core controller sets up crosspoint connections in the space switch core and sends control information (e.g. a connection grant) back to the input buffer modules to schedule the transmission of cells therefrom.
- This control information is passed through the control links 36 which are both functionally and physically separate from the data links. Synchronization between control and data links ensures the establishment and holding of connections in the space switch core in time with the passage of a train of cells.
- FIG 4 illustrates the input buffer module at input port A. Ports are bidirectional and hereinafter are labeled port A to port P.
- data at lower speed links from the access peripherals (channels 1-n) 50 are converted from serial to parallel ATM cells at a converter 52 and then multiplexed at a multiplexer 54 into a single parallel cell bus 56.
- the attached switch routing tags are read and cells are stored in a cell RAM 58 under the control of the buffer control circuit 60 according to the destination port identified by the switch routing tags.
- the stored ATM cells are transmitted serially through a data link 64 to a switch core (shown in detail in Figure 5) by a multiplexer 66.
- the switch routing tags contain a field which identifies the output buffer module.
- the buffer control circuit 60 uses this information to place the cell into the section of RAM assigned to the destination and to send a request for connection message to the space switch core over the control link 62.
- the input buffer module comprises a single or dual RAM with a cell wide data bus (busses).
- Many other configurations of the cell RAM are possible, such as a separate FIFO for each destination port or a common memory for all the ports.
- cells can be prioritized.
- the cell RAM can be configured to store cells according to their priority.
- the location FIFO for the particular destination port provides a sequence of RAM locations from which to send a train of cells.
- priority handling capability cells with high priority may be read before low priority cells. As cells are sent to the space switch core, their previous locations become available again and can be added back into the free list of the FIFO.
- the output scheduler 70 of the buffer control circuit in Figure 4 selects a train of cells for the same destination port.
- the input side of the input buffer module may be time divided (commutated) over a number of lower speed tributary links from the transport peripherals, just like a common memory switch.
- the FIFO manager 68 is directly responsive to the routing tags of cells (which in turn are derived from ATM cell headers) received in each cell cycle, in a very similar manner as the FIFO manager of a common memory switch. The only difference is that here the division into N pages or FIFOs is according to destination ports, i.e. ports on the space switch output.
- the space switch core 72 includes an optical shuffle 74, 16 1:16 optical splitters 74, and 16 opto-electrical data selectors 78.
- the optical shuffle distributes data cells from one input port to all 16 output ports, and the opto- electrical data selector at an output port chooses one desired input under control of the core controller 80.
- the core controller receives all request messages on control links 62, arbitrates between them, and returns grant messages when the appropriate connection is set up.
- the input buffer control 60 ( Figure 4) receives a corresponding grant some time after the request was sent. In the meantime, additional ATM cells for the same port destination may have arrived. Thus, when the grant message arrives, several ATM cells can be sent in a single train over the current connection. Therefore, given a large enough cell RAM, the core control can be arbitrarily slow in processing requests and granting connections for the transmission of trains of ATM cells. This simply results in longer bursts and longer waiting times. The algorithm, of course, runs fast enough to avoid exceeding the permissible delay for the switched traffic. Buffer space in the cell RAMs is also dimensioned appropriately to hold the waiting cells without overflowing.
- control and data links are separate and synchronization between them enables the use of the burst mode of switching in that
- connection is set up for a duration much longer than one cell period
- the value of [i] is sent to the core controller over the control link.
- the core controller accumulates that information. As a result, it eventually sets up a space switch connection for [i], and simultaneously sends a command to the buffer to release the corresponding batch of cells.
- M cells for destination [i] may have been reported and accumulated.
- the cells in the cell RAM can be counted by a simple counter and a simple counter value M can be sent to the core control circuit.
- the command "send M cells from buffer [i]" is received by the buffer output scheduler, it will start to transmit cells from [i] on the high speed link to the space switch core until M cells are sent or the FIFO is empty. It will then send idle cells until a new command is received from the space switch core.
- Figure 8 shows in detail the space switch core in a single rack unit and includes 16 port cards 90, and core common equipment 92. Connections are identified by circled numbers.
- the core common equipment contains the passive 256 x256 optical shuffle interconnect 98 and a common controller circuit 100. Many optical arrangements can be used to shuffle a plurality of optical inputs among a plurality of optical outputs, e.g. a fiber optic bundle etc.
- Figure 9 illustrates an alternative arrangement of the optical shuffle.
- the electrical bus between the port cards and the common controller circuit forms the control bus 102.
- each of the 16 horizontal slabs 110 forms a planar 1:16 splitter which converts the single fiber input of one port into parallel ribbons 112 of 16 fibers each.
- Each ribbon is connected back to the optical selectors of each port card.
- the optical data inputs on a data link are each split (replicated optically) 16 ways and connected to the optical shuffle.
- Each selector has 16 optical inputs and receives all data inputs from all the port cards.
- This optical arrangement provides the necessary shuffle and crossover of the entire switch bandwidth in an efficient, compact and low cost way and permits both the high speed of the optical signal to keep the number of signal paths low, and the ability to route densely packed optical paths without mutual interference.
- the electro-optical selector is a hybrid assembly on a silicon substrate with 16 optical inputs 120, one optical output 122, and a 4-bit electronic control input 124.
- the optical fibers are physically attached in grooves etched into the silicon which provide accurate alignment to a PIN diode array 126.
- the PIN diodes are standard high speed diodes and are fabricated as arrays on 250 ⁇ m spacing to match the spacing of the fibers.
- the diodes are bonded across to the GaAs device which contains amplifiers 128 and selection circuitry 130, driven by a clocked or non-clocked 4-bit control input.
- the switch control signal chooses one electrical signal, which corresponds to one of 16 optical inputs, to pass to a laser driver circuit 132 which further amplifies and drives the separately packaged laser 134.
- An output fiber 122 is coupled to the laser 134. If necessary, the selected data signal is reclocked at a data reclock 136 before leaving the GaAs device. Reclocking may be necessary to reduce edge noise, but without it the circuit would handle a wider range of bit rates.
- FIG 11 illustrates schematically the space core control.
- the purpose of the core control is to receive buffer reports from the input buffers and set the space core connection configuration accordingly.
- the core control also sends this same information back to the input buffers to trigger (grant) the release of the ATM bursts that will match the connection setup.
- This task is accomplished with a state machine in the common core control chip (CCCC) 150 through the control bus 152 which is able to absorb a peak of 16 connection requests per cell period, that is, one from each input buffer, and issue at least one connection setup and one grant per cell period.
- CCCC can be made in a single VLSI but pinout and bus rates can be traded off, up to 100 MHz, to optimize cost.
- PIC per-port control interface chip
- Control information is exchanged with the input buffer through RLIC (Receive Line Interface Chip) and TLIC (Transmit Line Interface Chip). Data pass through the data selector under the control of the PIC.
- the PIC isolates the control links from the CCCC to filter the buffer report information and removes all redundant information from the stream of request messages. It keeps track of whether a request has been queued in the CCCC and keeps updated on the current buffer fill of each of the ATM queues in the input buffer.
- control bus interface As a control bus interface, it allows a narrow bus to distribute the CCCC output signals, i.e. the XY addresses of the current connection set-up.
- the PIC reads this information on the control bus and drives its associated data bus selector.
- At each input buffer only the first request of each kind is passed on to the CCCC until a connection is made to grant the request. Subsequent requests to the same destination are counted in the PIC, and are later determined for the size of the burst that will be granted to the input buffer.
- the PIC receives buffer requests consisting of destination port id (4 bits significant), and current buffer fill (8 or more bits). It ignores null requests and stores the current buffer fill in a 16 register file. Further, there is a 16 by 1 bit flag register to store whether there is an outstanding request to the CCCC. Upon receipt of a buffer request, and if there is no outstanding request, that request (id only) is transmitted to the CCCC over the control bus (functionally shown by arrows 156), and the flag is set.
- the PIC When the CCCC assigns a connection, the PIC receives this indication over the control bus, clears the request flag, and sends a buffer grant message back to the input buffer through TLIC; this message contains the destination port id, and a copy of the current buffer fill, as stored in the PIC register file.
- the CCCC may receive up to 16 new buffer requests per cell tick (common chip processing cycle), but as it will only grant 1 or n per cycle, it will reach an equilibrium of receiving 1 or n per cycle as well.
- each new request is put at the end of an internal queue of requests which is a list of all outstanding requests in sequence, with the oldest request at the head. Since there are up to 256 possible connections, the maximum size queue is 256. This queue is processed as fast as feasible to satisfy connection requests, i.e. 1 or n per cell tick. To satisfy a request, at each cell tick the queue is scanned backwards from the head until a request is found for which both the source and the destination are free.
- a CAM (Content Addressable Memory) implementation of the queue scans this asynchronously and very rapidly.
- the corresponding request is then granted back to the PIC on the source card which PIC will copy the current buffer fill to a counter and ultimately to the requesting buffer, along with the accumulated count, to enable a burst of specified size.
- a record (flag registers) of both source and destination busy status is updated.
- the PIC sets the data selector.
- the counter on the source card is decremented by 1 for each cell tick. When it reaches 0, the agreed number of cells have been transmitted and the connection can be broken, i.e. the associated source and destination port busy flags in the CCCC must be cleared.
- This event can occur simultaneously in all PICs, and is reported back to the CCCC via the control bus or dedicated (1 per PIC) lines.
- the counters could also be located in the CCCC, requiring that the buffer fill value is reported from PIC to CCCC when the connection is set up.
- Figure 12 shows the core control mechanism according to another embodiment of the invention. Unlike the common core control which has been described thus far, in this implementation, control is distributed to all the ports and is communicated through two sets of buses (grant bus and request bus). The figure shows a pair of port control circuits 160 and 162 exchanging control through request bus 164 and grant bus 166. As stated earlier, the buffer control keeps track of the status of input FIFO for each destination port and generates status information for each destination. This information is transmitted to port control circuit 160 over the control link.
- FIFO notEmpty bits 168 In addition to a register file which stores the actual cell count per port (not shown), a single bit per port is set for each destination which has an outstanding request and is shown in the figure as FIFO notEmpty bits 168, indicating that FIFOs have data to send to their assigned destinations.
- a multiplexer 170 scans FIFO notEmpty bits and sends connection requests to its assigned request bus segment if it is not busy with a connection as shown by a source busy block 172.
- One port drives only one assigned bus segment but all the ports monitor all the segments by scanning with multiplexer 174.
- the destination port latches the address at 178, sends a 4-bit control signal to the space switch selector and returns a connection grant to the requesting port through its assigned segment of the grant bus.
- the requesting port receives the grant on the segment of the bus by scanning at multiplexer 180 and will recognize from which port the grant is received. If the requesting port is not busy with any other port , the destination address is latched at 182 and is sent to the buffer control for data transmission. During the subsequent data transfer, source 160 will not place another request on bus 164, and destination 162 will not respond to any requests until the predetermined count of cells has been transferred as monitored by a counter 184.
- the source status from the source busy block 172 on the source 160 is transmitted on a release bus (not shown).
- This bus is scanned by the destination 162 to determine when the predetermined count of cells have been transmitted.
- Figure 13 shows another embodiment of the invention in which WDM (wavelength division multiplexing) is used for transmission of the control and data streams through a single optical fiber, avoiding the problem of time skew between them.
- WDM wavelength division multiplexing
- the frequency distribution charts in Figures 14 and 15 illustrate the simulated behavior of the first algorithm described for one simple case of 70% occupancy of all links, random (Poisson) traffic, with two idle cells forced inserted after every burst
- the delay distribution in Figure 14 exhibits a characteristic square shape which is a result of the burst connection model: the last cell in a connection burst has to wait the shortest time, and the waiting time of the cells in the burst is, on the average, evenly distributed between the shortest (almost zero) and the longest service time of the request queue (about 26.0 microseconds here).
- the algorithm is designed to minimize not the average, but the maximum cell delay under heavy traffic.
- the duration (burst length) of each connection is distributed around a mean of 0.77 ⁇ sec in the same simulation.
Abstract
Description
Claims
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
EP95913837A EP0796546B1 (en) | 1994-12-08 | 1995-03-29 | Large capacity atm switch |
CA002204172A CA2204172C (en) | 1994-12-08 | 1995-03-29 | Large capacity atm switch |
JP8517208A JP2923693B2 (en) | 1994-12-08 | 1995-03-29 | Large capacity ATM switch |
DE69520229T DE69520229T2 (en) | 1994-12-08 | 1995-03-29 | ATM switchboard with higher capacity |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US08/352,405 | 1994-12-08 | ||
US08/352,405 US5475679A (en) | 1994-12-08 | 1994-12-08 | Large capacity ATM switch |
Publications (1)
Publication Number | Publication Date |
---|---|
WO1996018278A1 true WO1996018278A1 (en) | 1996-06-13 |
Family
ID=23384999
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CA1995/000184 WO1996018278A1 (en) | 1994-12-08 | 1995-03-29 | Large capacity atm switch |
Country Status (6)
Country | Link |
---|---|
US (1) | US5475679A (en) |
EP (1) | EP0796546B1 (en) |
JP (1) | JP2923693B2 (en) |
CA (1) | CA2204172C (en) |
DE (1) | DE69520229T2 (en) |
WO (1) | WO1996018278A1 (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2003056771A1 (en) * | 2001-12-27 | 2003-07-10 | Telefonaktiebolaget Lm Ericsson (Publ) | Method for reducing the amount of needed memory in a tdm switch system |
US7873056B2 (en) | 2004-11-18 | 2011-01-18 | Nec Corporation | Switch device, switching method and switch control program |
Families Citing this family (82)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
GB2288096B (en) * | 1994-03-23 | 1999-04-28 | Roke Manor Research | Apparatus and method of processing bandwidth requirements in an ATM switch |
SE516731C2 (en) * | 1994-06-10 | 2002-02-19 | Ericsson Telefon Ab L M | ATM-adapted three-stage selector unit |
FR2729040B1 (en) * | 1994-12-29 | 1997-01-31 | Alcatel Nv | PHOTON SWITCHING MATRIX |
US6055238A (en) * | 1995-03-17 | 2000-04-25 | Hitachi, Ltd. | Packet terminal and packet communication network |
GB2300086B (en) * | 1995-04-18 | 1999-08-04 | Northern Telecom Ltd | Switching arrangement |
US5623495A (en) * | 1995-06-15 | 1997-04-22 | Lucent Technologies Inc. | Portable base station architecture for an AD-HOC ATM lan |
JP3434642B2 (en) * | 1995-07-07 | 2003-08-11 | 株式会社東芝 | Packet scheduling device |
AU6502696A (en) * | 1995-07-19 | 1997-02-18 | Fujitsu Limited | Priority arbitration for point-to-point and multipoint transmission |
US5996019A (en) | 1995-07-19 | 1999-11-30 | Fujitsu Network Communications, Inc. | Network link access scheduling using a plurality of prioritized lists containing queue identifiers |
WO1997010656A1 (en) | 1995-09-14 | 1997-03-20 | Fujitsu Network Communications, Inc. | Transmitter controlled flow control for buffer allocation in wide area atm networks |
US5818840A (en) * | 1995-12-14 | 1998-10-06 | Time Warner Entertainment Co. L.P. | Asymmetric ATM switch |
US5940377A (en) * | 1995-12-21 | 1999-08-17 | Kabushiki Kaisha Toshiba | Asynchronous transfer mode switch including cell counter verification circuitry |
WO1997026737A1 (en) | 1996-01-16 | 1997-07-24 | Fujitsu Limited | A reliable and flexible multicast mechanism for atm networks |
JPH09247176A (en) * | 1996-03-11 | 1997-09-19 | Hitachi Ltd | Asynchronous transfer mode exchange system |
GB2311180B (en) * | 1996-03-13 | 2000-03-22 | Northern Telecom Ltd | Switch architecture |
JP3036586B2 (en) * | 1996-05-09 | 2000-04-24 | 日本電信電話株式会社 | ATM switch |
US6442172B1 (en) | 1996-07-11 | 2002-08-27 | Alcatel Internetworking, Inc. | Input buffering and queue status-based output control for a digital traffic switch |
US5768257A (en) * | 1996-07-11 | 1998-06-16 | Xylan Corporation | Input buffering/output control for a digital traffic switch |
US5748905A (en) | 1996-08-30 | 1998-05-05 | Fujitsu Network Communications, Inc. | Frame classification using classification keys |
US6400715B1 (en) * | 1996-09-18 | 2002-06-04 | Texas Instruments Incorporated | Network address matching circuit and method |
US5777529A (en) * | 1996-10-10 | 1998-07-07 | Northern Telecom Limited | Integrated circuit assembly for distributed broadcasting of high speed chip input signals |
FR2754964B1 (en) * | 1996-10-18 | 1999-07-16 | Thomson Csf | DEVICE WITH ECONOMIC DISTRIBUTED ARCHITECTURE FOR SWITCHING ATM CELLS IN A LOCAL AREA NETWORK, WITH BREAKDOWN TOLERANCE OF BREWING AND MANAGEMENT FUNCTIONS |
DE69733741T2 (en) * | 1996-12-20 | 2006-04-20 | International Business Machines Corp. | switching system |
DE69737676T2 (en) * | 1996-12-20 | 2008-01-10 | International Business Machines Corp. | Switching system for connecting line adapters containing distributed elements |
DE69734968T2 (en) * | 1996-12-20 | 2006-07-27 | International Business Machines Corp. | Distributed element switching system for connection to line adjusters and with multiple transmission capability |
US5978370A (en) * | 1997-01-13 | 1999-11-02 | At&Tcorp | Circuit-switched switching system |
GB2321563B (en) * | 1997-01-24 | 1999-04-21 | Plessey Telecomm | Burst mode wavelength manager |
US6118776A (en) * | 1997-02-18 | 2000-09-12 | Vixel Corporation | Methods and apparatus for fiber channel interconnection of private loop devices |
US6031821A (en) * | 1997-08-19 | 2000-02-29 | Advanced Micro Devices, Inc. | Apparatus and method for generating a pause frame in a buffered distributor based on lengths of data packets distributed according to a round robin repeater arbitration |
US6192046B1 (en) * | 1997-08-28 | 2001-02-20 | Ascend Communications, Inc. | Apparatus and method for efficiently transferring ATM cells across a backplane in a network switch |
US6295299B1 (en) * | 1997-08-29 | 2001-09-25 | Extreme Networks, Inc. | Data path architecture for a LAN switch |
US6324165B1 (en) * | 1997-09-05 | 2001-11-27 | Nec Usa, Inc. | Large capacity, multiclass core ATM switch architecture |
DE19751559A1 (en) * | 1997-11-20 | 1999-05-27 | Siemens Ag | ATM exchange arrangement for high data transfer speed |
DE19751560A1 (en) * | 1997-11-20 | 1999-05-27 | Siemens Ag | ATM multiplexer unit for high data transfer speed |
US6118792A (en) * | 1997-11-21 | 2000-09-12 | Nortel Networks Corporation | Method and apparatus for a flexible access rate common-memory packet switch |
AU2886599A (en) * | 1998-03-04 | 1999-09-20 | Ciena Corporation | Optical shuffle network interconnection |
JPH11275614A (en) * | 1998-03-26 | 1999-10-08 | Nec Corp | Photo switching device |
US6252849B1 (en) * | 1998-06-30 | 2001-06-26 | Sun Microsystems, Inc. | Flow control using output port buffer allocation |
JP3141850B2 (en) * | 1998-07-10 | 2001-03-07 | 日本電気株式会社 | Time division switching device, time division switching method, and recording medium |
FR2782430B1 (en) * | 1998-08-13 | 2004-05-28 | Bull Sa | INTERCONNECTION METHOD AND INTERFACE IMPLEMENTING HIGH-SPEED LINKS |
US6535313B1 (en) * | 1998-10-16 | 2003-03-18 | Lucent Technologies Inc. | Dynamically assignable optical signal access control apparatus |
US6621818B1 (en) * | 1998-10-19 | 2003-09-16 | Texas Instruments Incorporated | Ring configuration for network switches |
JP3866425B2 (en) * | 1998-11-12 | 2007-01-10 | 株式会社日立コミュニケーションテクノロジー | Packet switch |
US6721271B1 (en) | 1999-02-04 | 2004-04-13 | Nortel Networks Limited | Rate-controlled multi-class high-capacity packet switch |
US6647011B1 (en) * | 1999-02-22 | 2003-11-11 | Marconi Communications, Inc. | Method and system for switching using an arbitrator |
US6486983B1 (en) | 1999-12-30 | 2002-11-26 | Nortel Networks Limited | Agile optical-core distributed packet switch |
CA2303605A1 (en) * | 1999-04-06 | 2000-10-06 | Nortel Networks Corporation | Self-configuring distributed switch |
US6570872B1 (en) * | 1999-04-06 | 2003-05-27 | Nortel Networks Limited | Self-configuring distributed switch |
US6934471B1 (en) * | 1999-06-03 | 2005-08-23 | Avici Systems, Inc. | Photonic switch using time-slot interchange |
GB2352105B (en) * | 1999-07-15 | 2001-09-26 | Marconi Comm Ltd | Optical system |
JP2001045030A (en) * | 1999-07-29 | 2001-02-16 | Nec Corp | Connection controller |
US6891424B1 (en) * | 1999-09-29 | 2005-05-10 | Telasic Communications, Inc. | Monolithic payload IF switch |
JP2001127766A (en) * | 1999-10-25 | 2001-05-11 | Toshiba Corp | Line interface and packet exchange |
US6625148B1 (en) * | 1999-11-18 | 2003-09-23 | 3Com Corporation | Self routing interconnect cross-switch |
US6594714B1 (en) * | 2000-05-01 | 2003-07-15 | Hewlett-Packard Development Company, L.P. | Reconfigurable FIFO interface to support multiple channels in bundled agent configurations |
US6853635B1 (en) * | 2000-07-24 | 2005-02-08 | Nortel Networks Limited | Multi-dimensional lattice network |
IL137732A (en) * | 2000-08-07 | 2006-12-31 | Crosslight Photonics Ltd | Characterization of multiple section semiconductor lasers |
US7283547B1 (en) * | 2000-08-11 | 2007-10-16 | Ericsson Ab | Switch fabrics logical synchronization utilizing propagation lockdown |
IL139618A0 (en) * | 2000-11-12 | 2002-02-10 | Crosslight Photonics Ltd | Wavelenght tuning optimization of semiconductor lasers |
DE10118295A1 (en) * | 2001-04-12 | 2002-10-17 | Alcatel Sa | Optical crossconnect |
EP1393498B1 (en) * | 2001-05-17 | 2011-01-05 | Alcatel Lucent | Distributed shared memory packet switch |
US7167480B1 (en) * | 2001-06-21 | 2007-01-23 | Lighthouse Capital Partners Iv, Lp | Multi-service data transport architecture |
US7233590B2 (en) * | 2001-07-06 | 2007-06-19 | Nortel Networks Limited | Switched channel-band network |
US6836815B1 (en) * | 2001-07-11 | 2004-12-28 | Pasternak Solutions Llc | Layered crossbar for interconnection of multiple processors and shared memories |
DE60232519D1 (en) * | 2001-07-17 | 2009-07-16 | Alcatel Internetworking Inc | INBAND MESSAGE SYNCHRONIZATION FOR A DISTRIBUTED PACKAGE TRANSMISSION WITH A SHARED MEMORY |
US6914901B1 (en) * | 2001-12-05 | 2005-07-05 | Cisco Technology, Inc. | System and method for communicating using multiple memory banks |
KR100440574B1 (en) * | 2001-12-26 | 2004-07-21 | 한국전자통신연구원 | Variable Length Packet Switch |
AU2003207955A1 (en) * | 2002-01-10 | 2003-07-24 | Xlight Photonics Ltd. | Multiple port arbiter and switching systems |
US7394806B2 (en) * | 2002-04-11 | 2008-07-01 | Nortel Networks Limited | Distributed space-time-space switch |
DE10217856B4 (en) * | 2002-04-22 | 2006-11-02 | Infineon Technologies Ag | Telecommunications network and method for transmitting data in a telecommunication network |
US7272309B1 (en) * | 2002-05-08 | 2007-09-18 | Yotta Networks, Llc | System and method of routing data at a photonic core |
US7251416B1 (en) * | 2002-05-08 | 2007-07-31 | Yotta Networks, Llc | Container based crossconnect |
US7701932B2 (en) * | 2003-06-18 | 2010-04-20 | Motorola, Inc. | Distributed automatically-configuring switching system |
KR100473814B1 (en) * | 2003-07-16 | 2005-03-14 | 한국전자통신연구원 | Duplicated system and method using a serial to deserialize |
US20050177660A1 (en) * | 2004-02-05 | 2005-08-11 | Rajesh Mamidwar | Method and system for merged rate-smoothing buffer with burst buffer |
US8279759B1 (en) * | 2005-03-07 | 2012-10-02 | Verizon Services Corp. | Protocol inter-worked ping mechanism |
WO2007127081A2 (en) * | 2006-04-21 | 2007-11-08 | Pesa Switching Systems, Inc. | Distributed routing system and method |
US9729946B2 (en) * | 2009-04-03 | 2017-08-08 | Infinera Corporation | High-capacity switch |
US9325604B2 (en) * | 2013-03-15 | 2016-04-26 | Plexxi Inc. | System and method for data center optical connection |
US9736556B2 (en) * | 2015-09-10 | 2017-08-15 | Equinix, Inc. | Automated fiber cross-connect service within a multi-tenant interconnection facility |
US20170373982A1 (en) * | 2016-06-24 | 2017-12-28 | Huawei Technologies Co., Ltd. | System and method for mtu size reduction in a packet network |
KR20200131905A (en) * | 2018-04-13 | 2020-11-24 | 콤스코프 테크놀로지스, 엘엘씨 | Configurable wide area distributed antenna system |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO1984000266A1 (en) * | 1982-06-25 | 1984-01-19 | Western Electric Co | End-to-end information memory arrangement in a line controller |
US5126999A (en) * | 1989-04-20 | 1992-06-30 | Northern Telecom Limited | Method and apparatus for input-buffered asynchronous transfer mode switching |
EP0602382A2 (en) * | 1992-12-18 | 1994-06-22 | International Business Machines Corporation | A two-dimensional round-robin scheduling mechanism for switches with multiple input queues |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CA1065977A (en) * | 1977-05-09 | 1979-11-06 | Real Gagnier | Switching network for a pcm tdm system |
US4701907C1 (en) * | 1986-02-03 | 2002-08-27 | Collins Mary | Dynamically reconfigurable time-space-time digital switch and network |
JP2595025B2 (en) * | 1988-03-18 | 1997-03-26 | 株式会社日立製作所 | High-speed packet switching equipment using space division type switches |
JP2964457B2 (en) * | 1989-12-05 | 1999-10-18 | 株式会社日立製作所 | Communication processing device |
-
1994
- 1994-12-08 US US08/352,405 patent/US5475679A/en not_active Expired - Fee Related
-
1995
- 1995-03-29 EP EP95913837A patent/EP0796546B1/en not_active Expired - Lifetime
- 1995-03-29 WO PCT/CA1995/000184 patent/WO1996018278A1/en active IP Right Grant
- 1995-03-29 CA CA002204172A patent/CA2204172C/en not_active Expired - Fee Related
- 1995-03-29 JP JP8517208A patent/JP2923693B2/en not_active Expired - Fee Related
- 1995-03-29 DE DE69520229T patent/DE69520229T2/en not_active Expired - Fee Related
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO1984000266A1 (en) * | 1982-06-25 | 1984-01-19 | Western Electric Co | End-to-end information memory arrangement in a line controller |
US5126999A (en) * | 1989-04-20 | 1992-06-30 | Northern Telecom Limited | Method and apparatus for input-buffered asynchronous transfer mode switching |
EP0602382A2 (en) * | 1992-12-18 | 1994-06-22 | International Business Machines Corporation | A two-dimensional round-robin scheduling mechanism for switches with multiple input queues |
Non-Patent Citations (1)
Title |
---|
M. DECINA ET AL.: "Datagram switching networks for broadband data communications", IEEE GLOBAL TELECOMMUNICATIONS CONFERENCE '92, 6 December 1992 (1992-12-06), FLORIDA, US, pages 104 - 110, XP000357769 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2003056771A1 (en) * | 2001-12-27 | 2003-07-10 | Telefonaktiebolaget Lm Ericsson (Publ) | Method for reducing the amount of needed memory in a tdm switch system |
US7483425B2 (en) | 2001-12-27 | 2009-01-27 | Telefonaktiebolaget L M Ericsson (Publ) | Method for reducing the amount of needed memory in a TDM switch system |
US7873056B2 (en) | 2004-11-18 | 2011-01-18 | Nec Corporation | Switch device, switching method and switch control program |
Also Published As
Publication number | Publication date |
---|---|
DE69520229T2 (en) | 2001-06-28 |
US5475679A (en) | 1995-12-12 |
EP0796546A1 (en) | 1997-09-24 |
JP2923693B2 (en) | 1999-07-26 |
DE69520229D1 (en) | 2001-04-05 |
CA2204172C (en) | 2001-08-14 |
CA2204172A1 (en) | 1996-06-13 |
EP0796546B1 (en) | 2001-02-28 |
JPH10511820A (en) | 1998-11-10 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US5475679A (en) | Large capacity ATM switch | |
US5361255A (en) | Method and apparatus for a high speed asynchronous transfer mode switch | |
Rojas-Cessa et al. | CIXB-1: Combined input-one-cell-crosspoint buffered switch | |
EP0705006B1 (en) | ATM communication system for statistical multiplexing of cells | |
US7023841B2 (en) | Three-stage switch fabric with buffered crossbar devices | |
US7173931B2 (en) | Scheduling the dispatch of cells in multistage switches | |
US5844887A (en) | ATM switching fabric | |
EP0300061B1 (en) | Self-routing switching system | |
JP3074353B2 (en) | Large capacity ATM switch | |
CA2224606C (en) | A distributed buffering system for atm switches | |
EP0858192A2 (en) | An ATM switching arrangement | |
EP0590877B1 (en) | Multistage optical packet distribution network with bypass | |
US6768717B1 (en) | Apparatus and method for traffic shaping in a network switch | |
US5499239A (en) | Large capacity modular ATM switch | |
EP0590865B1 (en) | Multistage growable optical packet switching arrangement with bypass | |
US6101190A (en) | Switching system and method for asynchronous transfer mode exchange for multimedia service | |
US5285445A (en) | Switching network and switching network control for a switching node in a wideband transmission system | |
Chao et al. | Design and implementation of Abacus switch: A scalable multicast ATM switch | |
US7103056B2 (en) | Scheduling the dispatch of cells in multistage switches using a hierarchical arbitration scheme for matching non-empty virtual output queues of a module with outgoing links of the module | |
US20030012214A1 (en) | Hybrid time switch as a rotator tandem | |
JPH08251235A (en) | Tera-bit/see packet switch | |
Obara et al. | High speed transport processor for broad-band burst transport system | |
JP4504606B2 (en) | Apparatus and method for shaping traffic in a network switch | |
JP2546490B2 (en) | Switching system | |
JP3075187B2 (en) | ATM switch |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AK | Designated states |
Kind code of ref document: A1 Designated state(s): CA JP |
|
AL | Designated countries for regional patents |
Kind code of ref document: A1 Designated state(s): AT BE CH DE DK ES FR GB GR IE IT LU MC NL PT SE |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
DFPE | Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101) | ||
ENP | Entry into the national phase |
Ref document number: 2204172 Country of ref document: CA Kind code of ref document: A Ref document number: 2204172 Country of ref document: CA |
|
WWE | Wipo information: entry into national phase |
Ref document number: 1995913837 Country of ref document: EP |
|
WWP | Wipo information: published in national office |
Ref document number: 1995913837 Country of ref document: EP |
|
WWG | Wipo information: grant in national office |
Ref document number: 1995913837 Country of ref document: EP |