Embedding rings on a toroid computer network

A method of generating a set of programs to be executed in parallel on a computer comprising a plurality of interconnected processing nodes arranged in a toroid configuration with multiple layers arranged along an axis;

• each layer comprising a plurality of processing nodes connected in a ring in a non-axial plane by an intralayer respective set of links between each pair of neighbouring processing nodes, the links in each set adapted to operate simultaneously;
• wherein the processing nodes in each layer are connected to respective corresponding nodes in each adjacent layer by an interlayer link to form respective rings along the axis;
• the method comprising:
• generating a first data transmission instruction for a first program, wherein the first data transmission instruction comprises a first link identifier which defines a first outgoing link from a first processing node on which data is to be transmitted, wherein the first program is generated to transmit the data with one of a symmetric or asymmetric bandwidth utilisation;
• generating a second data transmission instruction for a second program, wherein the second data transmission instruction comprises a second link identifier which defines a second outgoing link from a second processing node on which data is to be transmitted;
• and
• determining the first link identifier and the second link identifier to transmit data around a plurality of embedded one-dimensional logical paths in such a manner that the plurality of embedded one-dimensional logical paths operate simultaneously, each logical path using all processing nodes of the computer in sequence.

A method of executing a set of programs in parallel on a computer comprising a plurality of interconnected processing nodes arranged in a toroid configuration with multiple layers arranged along an axis;

• each layer comprising a plurality of processing nodes connected in a ring in a non-axial plane by an intralayer respective set of links between each pair of neighbouring processing nodes, the links in each set adapted to operate simultaneously;
• wherein processing nodes in each layer are connected to respective corresponding nodes in each adjacent layer by an interlayer link to form respective rings in a second dimension along the axis;
• the method comprising:
• executing a first data transmission instruction in a first program, wherein the first data transmission instruction comprises a first link identifier which defines a first outgoing link on a first processing node on which data is to be transmitted;
• executing a second data transmission instruction in a second program, wherein the second data transmission instruction comprises a second link identifier which defines a second outgoing link on a first processing node on which data is to be transmitted;
• and
• the first link identifier and the second link identifier having been determined to transmit data around each of a plurality of embedded one-dimensional logical paths formed by respective sets of the processing nodes and the intralayer and interlayer links such that the plurality of embedded one-dimensional logical paths operate simultaneously, each one-dimensional logical path using all processing nodes of the computer in sequence, wherein the data is transmitted with a symmetric bandwidth utilisation or an asymmetric bandwidth utilisation.