Abstract: A programmably sliceable switch-fabric unit (PSSU) and methods of use are disclosed. An N×N? switch matrix is programmably made to operate as if it were a plurality of S×S? virtual switch slices, where S<N and S?<N?. Ingressing requests each specify an egress path (unicast mode) or plural egress paths (multicast mode) in terms of one or more relative egress port numbers. A request translator converts relative egress port numbers into absolute egress port numbers by determining what virtual slice each request belongs to. The translated egress requests are handed off to an arbitration and/or scheduling mechanism for further processing. If the translated request is granted, the corresponding payload egresses through the actual egress port(s) which the translated request asked for.

Abstract: In a system having independently-clocked job-performing circuits (e.g., payload processors) and independently-clocked job-ordering circuits (e.g., request and payload suppliers), coordinating mechanisms are provided for coordinating exchanges between the independently-clocked circuits. The coordinating mechanisms include those that use transmitted time-stamps for scheduling contention-free performances within the job-performing circuits of requested jobs. The coordinating mechanisms additionally or alternatively include static and dynamic rate constraining means that are configured to prevent a faster-clocked one of the independently-clocked circuits from overwhelming a more slowly-clocked other of the independently-clocked circuits. In one implementation, independently-clocked telecommunication-shelves house a distributed set of line cards and switch cards.

Abstract: A method and apparatus for providing programmable memory functions for bi-directional cell traffic in a switch platform are provided, wherein a parameterized bi-directional FIFO unit controls cell traffic in a switch platform using a first and a second unidirectional FIFO buffer. The first and second unidirectional FIFO buffers each comprise asynchronous read and write ports. A cell size and a word size of the first and second unidirectional FIFO buffers are programmable. The bi-directional FIFO unit is coupled to write at least one cell from and read at least one cell to at least one asynchronous transfer mode (ATM) interface, at least one frame relay interface, at least one voice interface, and at least one data interface. As such, the first unidirectional FIFO buffer is coupled to write at least one cell from, and the second unidirectional FIFO buffer is coupled to read at least one cell to an ATM interface, a frame relay interface, a voice interface, and a data interface.

Abstract: A synchronization state for a local clock generating circuit of a first of a number of components of a distributed system is maintained according to a number of local clock cycles recorded between successive occurrences of a global synchronization signal provided to the components within the distributed system. The local clock generating circuit may enters the synchronization state only after observing a predetermined number of occurrences of successive local clock cycles between instances of the global synchronization signal. The local clock generating circuit continues to provide local control signals for the first of the components at time instants corresponding to the number of local clock cycles even after an instance of the global synchronization signal is observed at a time instant corresponding to one local clock cycle more or less than the number of local clock cycles.

Abstract: A method of managing a network switch. The method having the first step of detecting a status of a set of physical ports on an interface card in the network switch. Then, determining if the status is in a first state that indicates that all physical ports in the interface card are inaccessible. If the status is in the first state, then accepting all traffic for the set of physical ports. Also disclosed is an apparatus for performing the method.

Abstract: An apparatus and method for reducing congestion in a network switching node. A congestion mask that indicates which of a plurality of destination ports in a network switching node are congested is generated and combined with a destination port field included in a packet in a multicast queue of the switching node to mask destination port designations in the destination port field that are indicated by the congestion mask to be congested if a drop eligibility field within the packet indicates that destination port designations are permitted to be masked. The packet is dequeued from the multicast queue to be forwarded to destination ports in the network switching node indicated by destination port designations in the destination port field not masked by the combination of the congestion mask with the destination port field.

Abstract: A communication interface is described to align at a destination data transmitted through different channels before that data is read out. The communication interface includes a receiver circuit that has a plurality of buffers. Each buffer is coupled to a corresponding channel to receive data therethrough. The communication interface also includes a control circuit, coupled to the plurality of buffers, to enable reading of data from the plurality of buffers when each of the plurality of buffers has received at least one unit of data.

Abstract: A method and apparatus for rate-based cell traffic arbitration in a switch are provided, wherein arbitration is provided between eight traffic sources in the form of eight cell bus service modules on the same cell bus. A cell bus controller (CBC) is programmed with an 8-bit Relative Service Delay (RSD) value for each of the eight service modules. The value for each RSD is calculated based on the bandwidths allotted for each service module. This RSD value determines the portion of the total bandwidth of the switch platform reserved for the respective service module. Furthermore, each service module uses an 8-bit Service Delay Accumulator (SDA) register. The SDA register of each service module is configured using an SDA value, wherein the SDA register keeps track of when each of the service modules should receive service.

Abstract: A method and apparatus for routing cells having different formats among service modules of a switch platform are provided. The cells are routed among service modules of a switch by a cell bus controller (CBC) using a first memory to convert an address having a first format into an address having a second format. The address having the first format is received in a header of a cell, and the address format comprises a 17-bit cell bus logical connection number of a destination port. The address having the second format is a 16-bit UDF used by a switch of the switch platform. The address having the first format is used to form a third address that is used to access the first memory. The data located at the third address of the first memory is a 16-bit UDF used to address the switch. A second memory is used to convert an address having the second format into an address having the first format. The address having the second format is used as a fourth address to access the second memory.

Abstract: A method and apparatus for providing cell bus management in a switch platform are provided. Each unidirectional FIFO buffer of a cell bus controller outputs a write port cell count from a write port. A cell count value is programmed at which the write port cell count is outputted. When the write port cell count indicates that the FIFO buffer can not accept additional data or cells, a master bidirectional FIFO unit ceases reading cells to a unidirectional FIFO buffer of the slave bidirectional FIFO unit in response to the write port cell count. Furthermore, the master bidirectional FIFO unit disables a corresponding switch from routing cells to the slave bidirectional FIFO unit in response to the write port cell count; the switch routes the cells to another of the slave bidirectional FIFO units.

Abstract: A method and apparatus for providing asynchronous memory functions for bi-directional cell traffic in a switch platform are provided, wherein a parameterized bi-directional FIFO unit controls cell traffic in a switch platform using a first and a second unidirectional FIFO buffer. The first and second unidirectional FIFO buffers each comprises asynchronous read and write ports. A cell size and a word size of the first and second unidirectional FIFO buffers are programmable. The bi-directional FIFO unit is coupled to write at least one cell from and read at least one cell to at lest one asynchronous transfer mode (ATM) interface, at least one frame relay interface, at least one voice interface, and at least one data interface. As such, the first unidirectional FIFO buffer is coupled to write at least one cell from, and the second unidirectional FIFO buffer is coupled to read at least one cell to an ATM interface, a frame relay interface, a voice interface, and a data interface.