Abstract A switch queue structure for one-network parallel processor systems minimizes chip count and reduces the possibility of deadlock which might otherwise occur with this type of switch structure. The switch queue structure comprises a plurality of input ports and a plurality of output ports equal in number to a number of processor/memory elements (PMEs) in a parallel processor system. A plurality of identical stages interconnect the plurality of input ports and the plurality of output ports. Each stage includes a plurality of first groups of first-in, first-out (FIFO) registers storing request messages, a plurality of second groups of first-in, first-out registers storing response messages, and a plurality of multiplexers. Each of the multiplexers have inputs connected to the FIFO registers in each of the first and second groups for routing responses to memory requests from addressed processor/memory elements to requesting processor/memory elements and for routing memory requests from processor/memory elements to addressed processor/memory elements.
The tightly coupled parallel processor system recited in claim 1 wherein said plurality of identical stages are composed of a plurality of 2×2 switches, each of said switches including four of said first-in, first-out registers of said first group and four of said first-in, first-out registers of said second group and two of said multiplexers, each of said multiplexers having two inputs connected to two of said first-in, first-out registers of said first group and two of said first-in, first-out registers of said second group.
Claim 2
Dependent Claim
40 words
The tightly coupled parallel processor system recited in claim 1 wherein inputs from said first group of first-in, first-out registers have a higher priority than inputs from said second group of first-in, first-out registers thereby reducing the possibility of deadlock.
Claim 3
Dependent Claim
28 words
The tightly coupled parallel processor system recited in claim 2 wherein priority within said first and second groups of first-in, first-out registers is assigned on a round-robin basis.
Claim 4
Dependent Claim
73 words
The tightly coupled parallel processor system recited in claim 3 wherein each switch further includes: two input buffers connected to buffer messages to two first-in, first-out registers of each of said first and second groups of first-in, first-out registers; and two output buffers connected to outputs of said two multiplexers to buffer messages from selected ones of said fist-in, first-out registers in each of said first and second groups of first-in, first-out registers.