Computing system framework and method for configuration thereof
A computing system framework and method for configuration thereof are provided. A plurality of processing modules are accessed. Each processing module includes a plurality of processing nodes and each processing node is associated with an intra-module port and an inter-module port. A plurality of intra-module networks are formed. Each intra-module network includes connections between at least a portion of the processing nodes in one of the processing modules via the associated intra-module ports. An enclosed shape of the processing modules is formed by connecting at one inter-module port on each processing module to one inter-module port on an adjacent processing modules. A cable is linked between one of the inter-module ports of one processing module of the enclosed shape to an inter-module port of another processing module of a different group of interconnected processing modules.
1. A computing system framework, comprising:
a plurality of processing modules, each processing module comprising a plurality of processing nodes and each processing node associated with an intra-module port and an inter-module port;
a plurality of intra-module networks, each intra-module network comprising connections between at least a portion of the processing nodes in one of the processing modules via the associated intra-module ports;
a closed shape of the processing modules represented as M processing modules and formed by connecting at least one inter-module port on each processing module to one inter-module port on an adjacent processing module;
a network switch;
an inter-module network configured to form two or more connections between the closed shape of the processing modules and the network switch, each connection comprising at least one of:
a connection between the network switch and one of the processing modules N on the closed shape and another processing module N+S on the closed shape, by connecting one unused inter-module port on the processing module N to the switch and by connecting another unused inter-module port on the processing module N+S to the network switch, wherein S is an integer divisor of M, N is [x, x+S, . . . , x+(M−S), increment by S], and xis [1, 2, . . . , S−1];
a connection between the network switch and one of the processing modules N on the closed shape and another processing module N+S−1 on the closed shape, by connecting one unused inter-module port on the processing module N to the network switch and by connecting another unused inter-module port on the processing module N+S−1 the network switch, wherein S is an integer divisor of M, N is [x, x+S, . . . , x+(M−S), increment by S], x is [1, 2, . . . , S−1]; and
a connection between the network switch and one of the processing modules N on the closed shape and another processing module N+S−r on the closed shape, by connecting one unused inter-module port on the processing module N to the network switch and by connecting another unused inter-module port on the processing module N+S−r to the network switch, wherein S is an integer divisor of M, N is [x, x+S, . . . , x+(M−S), increment by S], x is [1, 2, . . . , S−1], and r is [2, 3, . . . , S−1]; and
a cable linked between one of the inter-module ports of one processing module of the closed shape to an inter-module port of another processing module of a different group of interconnected processing modules, wherein the different group of interconnected processing module has a configuration that is same or different than the closed shape.
2. A computing system framework according to claim 1 , wherein each processing node is selected from the group comprising:
a processing element, a memory controller, a memory, a storage controller, a storage device, interfaces to intra-module and inter-module ports, and a virtual node.
3. A computing system framework according to claim 1 , wherein the closed shape comprises one of a bus, a ring, a star, a mesh, and a crossbar switch.
4. A computing system framework according to claim 1 , further comprising:
a transmission module to transfer the data packets from one of the processing nodes to another one of the processing nodes via the connections.
5. A computing system framework according to claim 4 , further comprising:
a monitoring module measuring a pattern of the data packet traffic; and
a modification module to amend the connections based upon the packet traffic pattern.
6. A computing system framework according to claim 1 , wherein the network switch is arranged in a center of the closed shape of processing modules.
7. A method for constructing a computing system framework, comprising:
accessing a plurality of processing modules, each processing module comprising a plurality of processing nodes and each processing node associated with an intra-module port and an inter-module port;
within each of one or more of the processing modules, connecting at least a portion of the processing nodes in that processing module via the associated intra-module ports;
forming a closed shape of the processing modules represented as M processing modules by connecting at least one inter-module port on each processing module to one inter-module port on an adjacent processing module;
forming two or more connections around between the closed shape and a network switch, comprising at least one of:
a connection between the network switch and one of the processing modules N on the closed shape and another processing module N+S on the closed shape, by connecting one unused inter-module port on the processing module N to the network switch and by connecting another unused inter-module port on the processing module N+S to the network switch, wherein S is an integer divisor of M, N is [x, x+S, . . . , x+(M−S), increment by S], and x is [1, 2, . . . , S−1];
a connection between the network switch and one of the processing modules N on the closed shape and another processing module N+S−1 on the closed shape, by connecting one unused inter-module port on the processing module N to the network switch and by connecting another unused inter-module port on the processing module N+S−1 to the network switch, wherein S is an integer divisor of M, N is [x, x+S, . . . , x+(M−S), increment by S], x is [1, 2, . . . , S−1]; and
a connection between the network switch and one of the processing modules N on the closed shape and another processing module N+S−r on the closed shape, by connecting one unused inter-module port on the processing module N to the network switch and by connecting another unused inter-module port on the processing module N+S−r to the network switch, wherein S is an integer divisor of M, N is [x, x+S, . . . , x+(M−S), increment by S], x is [1, 2, . . . , S−1], and r is [2, 3, . . . , S−1]; and
linking a cable between one of the inter-module ports of one processing module of the closed shape to an inter-module port of another processing module of a different group of interconnected processing modules, wherein the different group of interconnected processing module has a configuration that is same or different than the closed shape.
8. A method according to claim 7 , wherein each processing node is selected from the group comprising:
a processing element, a memory controller, a memory, a storage controller, a storage device, interfaces to intra-module and inter-module ports, and a virtual node.
9. A method according to claim 7 , wherein the closed shape comprises one of a bus, a ring, a star, a mesh, and a crossbar switch.
10. A method according to claim 7 , further comprising:
transferring the data packets from one of the processing nodes to another one of the processing nodes via the connections.
11. A method according to claim 10 , further comprising:
measuring a pattern of the data packet traffic; and
amending the connections based upon the packet traffic pattern.
12. A method according to claim 7 , wherein the network switch is arranged in a center of the closed shape of processing modules.