摘要:
The invention proposes a switching arrangement for transporting data packets which comprise a data packet destination information and a payload, to one or more output ports. The switching device is able to route the arriving data packets according to the data packet destination information, to at least one dedicated of the output ports. It comprises at each input port an input buffer with at least as many single input queues as there are output ports, and an input controller for each input port, serving for controlling the order of multiplexing the data packets from the input queues of the corresponding input buffer to the switching device. The total of input ports is divided up into several subsets of input ports. Each subset in the switching device has its separate output buffer for storing at addresses therein at least the payload of each data packet arriving at the input port. At least one set of as many output queues as the switching arrangement has output ports are arranged. In these output queues at least the address of each payload stored in the output buffer is stored, sorted according to the data packet destination information. The stored payload is then deliverable to its dedicated at least one output port under use of the stored addresses.
摘要:
The present invention relates to switching technology in computer networks and in particular to a method and system for switching information packets through a m input, n output device. According to the invention it is proposed to temporarily buffer said packets according to a new, self-explanatory, preferred a linear addressing scheme in which a respective buffer location of consecutive stream packets results from a respective self-explanatory, or linear, respectively, incrementation of a buffer pointer. Preferably, a matrix of FIFO storage elements (10,11,12,13) having an input and an output crossbar can be used for implementing input/output paralleling modes (ILP,OLP) and multiple lanes and achieving address input/output scaling up to a single cycle.
摘要:
The present invention relates to a buffered crossbar switch which provides a step of changing the size and/or number of queuing buffer entries to ensure optimum buffer memory usage independent of the size of data packets processed.
摘要:
The present invention relates to a buffered crossbar switch and its method of operation which provides a step of changing the size and/or number of queuing buffer entries to ensure optimum buffer memory usage independent of the size of data packets processed.
摘要:
Apparatus and method of operating a crossbar switch (1) having a control logic, an output port scheduler (2), n input ports (i—0, . . . , i—31) and m output ports (o—0, . . . , o—31), wherein information packets are routed from said n input ports to said m output ports, and wherein said output port scheduler (2) controls the sequence of packets output at said output ports (o—0, . . . , o—31). To ensure fairness regarding packet transfer/routing and the packet sequence, an input port number corresponding to the input port a new information packet is arriving at is stored in round-robin mode in a port number buffer (pnb—0). For output, said input port number is retrieved from said port number buffer (pnb—0) in round robin mode, too, and with this port number, address information regarding the packet is obtained from a control logic buffer of the crossbar switch (1).
摘要:
The present invention relates to a method of operating a crossbar switch (1) having a control logic (2) and n input ports (i—0, . . . , i_n-1) and m output ports (o—0, . . . , o_m-1), wherein information packets of p different priority levels are routed from said n input ports (i—0, . . . , i_n-1) to said m output ports (o—0, . . . , o_m-1). Within said control logic (2), a pool (CRA) of buffers (CRA—0, CRA—1, . . . ) is provided for each crosspoint (4) for temporarily storing address information related to said information packets.
摘要:
The present invention relates to a method of operating a buffered crossbar switch. The proposed method reduces power dissipation in a buffered crossbar switch by reducing the number of crossbar buffer write processes.
摘要:
An improved multiple bus system for a multiprocessor computer system is disclosed for a computer system having a multiple level cache memory structure. The system includes one or more logical busses each including two or more physical busses for coupling multiple processors to a memory unit. Each logical bus is coupled to a bus switching unit which in turn couples all of the processors in the multiprocessor system to a memory unit over the physical busses comprising the logical bus. The system further manages near end signal reception problems caused by multiple processors electrically interconnected over such a bus system.
摘要:
An electrical circuit for generating clock pulses for a multi-chip computer system which contains a clock generation chip and various logic circuit chips. The clock pulses used on the logic circuit chips are generated on the clock generation chip and are transferred to the logic circuit chips. For the generation of the clock pulses a so-called clock splitter circuit is provided on the clock generation circuit. This clock splitter generates two pulse strings out of a third pulse string which is derived from an oscillator. The clock splitter contains a number of gates and latches which have an impact on the throughput time of a pulse to run through the clock splitter, as well as on the skew of the two generated pulse strings. The invention provides an electrical circuit which has an improved throughput time and skew of the generated pulse strings.
摘要:
A high availability shared cache memory in a tightly coupled multiprocessor system provides an error self-recovery mechanism for errors in the associated cache directory or the shared cache itself. After an error in a congruence class of the cache is indicated by an error status register, self-recovery is accomplished by invalidating all the entries in the shared cache directory means of the accessed congruence class by resetting Valid bits to "0" and by setting the Parity bit to a correct value, wherein the request for data to the main memory is not cancelled.Multiple bit failures in the cached data are recovered by setting the Valid bit in the matching column to "0". The processor reissues the request for data, which is loaded into the processor's private cache and the shared cache as well. Further requests to this data by other processors are served by the shared cache.