Abstract: A multi-module switching system comprising at least two switching modules adapted for receiving data packets from at least one input adapter and transmitting the data packets to at least one output adapter, each of the switching modules including a shared buffer for buffering a portion of a data packet received from an input adapter and transmitting the portion to an output adapter. One of the switching modules is a master module receiving a portion of a data packet containing a packet header and sending control information contained therein serially to each other switching module as a slave module.

Abstract: A network processor dataflow chip and method for flexible dataflow are provided. The dataflow chip comprises a plurality of on-chip data transmission and scheduling circuit structures. The data transmission and scheduling circuit structures are selected responsive to indicators. Data transmission circuit structures may comprise selectable frame processing and data transmission functions. Selectable frame processing may comprise cut and paste, full dispatch and store and dispatch frame processing. Scheduling functions include full internal scheduling, calendar scheduling in communication with an external scheduler, and external calendar scheduling. In another aspect of the present invention, data transmission functions may comprise low latency and normal latency external processor interfaces for selectively providing privileged access to dataflow chip resources.

Abstract: A data switch is provided which routes fixed-size data packets from input ports to output ports, using shared memory which holds a copy of each packet in buffers. Output ports have a queue which contains pointers to buffers holding packets bound for that port. The number of shared memory buffers holding packets is compared to the number of buffer pointers in the output queues. In this way, a Multicast Index (MCI), a metric of the level of multicast traffic, is derived. The switch includes a Switch Core Adaptation Layer (SCAL) which has a multicast input queue. Because traffic is handled based on priority class P, a multicast threshold MCT(P), associated with the multicast input queue, is established per priority. While receiving traffic, the MCI is updated and, for each priority class in each SCAL, the MCI is compared to the MCT(P) to determine whether corresponding multicast traffic must be held.

Abstract: The invention permits an effective traffic flow control, down to all sub-ports, of a switch made of a N-port core switch fabric. Sub-ports concentrate traffic from lower speed lines to a switch fabric native port. In each sub-port adapter, when congestion is detected in the OUT leg, it is reported through the corresponding IN leg. Congestion is piggybacked over the incoming traffic entering the input port of the N-port core switching fabric and is broadcast so that all sub-ports become aware of the detected congestion in any of the sub-ports. Each sub-port adapter performs a checking of the congestion status of all the other sub-ports and acts to stop forwarding received traffic destined for congested sub-ports and holds further received traffic until the sub-ports are reported to be no longer congested. The full intrinsic performance of a N-port switch fabric is realized by concentrating, through port and sub-port adapters, the traffic of more than N independent lines.

Abstract: An apparatus and process for updating a sample time in a serial link which converts serial data in parallel data. A delay line stores multiple samples of at least two data bits received over the serial link. The contents of the delay line are matched so that they can be analyzed by a processor to determine an optimum sampling position in the delay line. The processor is programmed to analyze contents of the latch by creating a sample mask from a plurality of delay line samples. The sample mask identifies transition edges of first and second data bits within the delay line. The transition edges are validated with respect to the presence, for first and second initial sampling positions for the respective data bits. New sampling positions are determined from the validated edges, and the initial sampling positions are updated with sampling positions which have been determined from the new sampling positions.

Abstract: A method to resequence packets includes sequentially allocating in each source ingress adapter a packet rank to each packet received within each source ingress adapter. In each destination egress adapter, each ranked data packet is stored at a respective buffer address of an egress buffer. The respective buffer addresses of data packets received by a same source ingress adapter with a same priority level and switched through a same switching plane are linked in a same linked-list. The respective buffer addresses are preferably linked by their order of use in the egress buffer, and thus each linked-list is having a head list pointing to the oldest buffer address. The linked-lists are sorted into subsets including those linked-lists linking the respective buffer addresses of data packets received by a same source ingress adapter with a same priority level.

Abstract: A queue scheduling mechanism in a data packet transmission system, the data packet transmission system including a transmission device for transmitting data packets, a reception device for receiving the data packets, a set of queue devices respectively associated with a set of priorities each defined by a priority rank for storing each data packet transmitted by the transmission device into the queue device corresponding to its priority rank and a queue scheduler for reading, at each packet cycle, a packet in one of the queue devices determined by a normal priority preemption algorithm. The queue scheduling mechanism includes a credit device that provides, at each packet cycle, a value N defining the priority rank to be read by the queue scheduler from the queue device corresponding to the priority N instead of the queue device determined by the normal priority preemption algorithm.

Abstract: A queue scheduling mechanism in a data packet transmission system, the data packet transmission system including a transmission device for transmitting data packets, a reception device for receiving the data packets, a set of queue devices respectively associated with a set of priorities each defined by a priority rank for storing each data packet transmitted by the transmission device into the queue device corresponding to its priority rank, and a queue scheduler for reading, at each packet cycle, a packet in one of the queue devices determined by a normal priority preemption algorithm. The queue scheduling mechanism includes a credit device that provides at each packet cycle a value N defining the priority rank to be considered by the queue scheduler whereby a data packet is read by the queue scheduler from the queue device corresponding to the priority N instead of the queue device determined by the normal priority preemption algorithm.

Abstract: A multi-module switching system comprising at least two switching modules adapted for receiving data packets from at least one input adapter and transmitting the data packets to at least one output adapter, each of the switching modules including a shared buffer for buffering a portion of a data packet received from an input adapter and transmitting the portion to an output adapter. One of the switching modules is a master module receiving a portion of a data packet containing a packet header and sending control information contained therein serially to each other switching module as a slave module.

Abstract: A data switch is provided which routes fixed-size data packets from input ports to output ports, using shared memory which holds a copy of each packet in buffers. Output ports have a queue which contains pointers to buffers holding packets bound for that port. The number of shared memory buffers holding packets is compared to the number of buffer pointers in the output queues. In this way, a Multicast Index (MCI), a metric of the level of multicast traffic, is derived. The switch includes a Switch Core Adaptation Layer (SCAL) which has a multicast input queue. Because traffic is handled based on priority class P, a multicast threshold MCT(P), associated with the multicast input queue, is established per priority. While receiving traffic, the MCI is updated and, for each priority class in each SCAL, the MCI is compared to the MCT(P) to determine whether corresponding multicast traffic must be held.

Abstract: The invention permits an effective traffic flow control, down to all sub-ports, of a switch made of a N-port core switch fabric. Sub-ports concentrate traffic from lower speed lines to a switch fabric native port. In each sub-port adapter, when congestion is detected in the OUT leg, it is reported through the corresponding IN leg. Congestion is piggybacked over the incoming traffic entering the input port of the N-port core switching fabric and is broadcast so that all sub-ports become aware of the detected congestion in any of the sub-ports. Each sub-port adapter performs a checking of the congestion status of all the other sub-ports and acts to stop forwarding received traffic destined for congested sub-ports and holds further received traffic until the sub-ports are reported to be no longer congested. The full intrinsic performance of a N-port switch fabric is realized by concentrating, through port and sub-port adapters, the traffic of more than N independent lines.

Abstract: A network node supports switching fixed length information cells between a source unit and a destination unit. The invention uses two lookup tables called the active and the standby calendars for each of said output line. Each entry in the calendars represents the position of one cell in the output cell stream which will be sent onto said output lines. While the active calendar is controlling the network cell multiplexing onto a node output line, the standby calendar is updated at each traffic change by looking for a calendar free position for the new traffic starting with a targeted theoretical entry position. A Free Location Table (700) is used, storing the calendar organized by pages. The standby calendar updating starts with reading a Free Location Table page pointed at by the most significant bits of the theoretical entry. Should said page contain only ones (i.e., no free cell position) looking for the nearest page containing at least one zero (i.e.

Abstract: A line-adapter of a communications controller includes, for scanning the teleprocessing lines connected to it, cyclic scanning means FES exchanging information with the lines through a serial bidirectional link on which data and control informations are partitioned into frames and slots. Since both the FES and the serial link work with their own timings, an interface FESA is provided to adapt the FES scanning to the serial link structure. This FESA includes temporary storage means for storing on the one hand, data and control information transmitted from the LICs to the FES (10) through the inbound serial link, and on the other hand, data and control information transmitted from the FES to the LICs through the outbound serial link. The access of the FES, the outbound and inbound serial link to the storage means is time-shared and granted by an arbitration logic, according to the relative priorities of operation of said elements within the line-adapter of the communications controller.