Computing system framework and method for configuration thereof
A computing system framework and method for configuration thereof are provided. A plurality of processing modules is accessed. Each processing module includes a plurality of processing nodes and each processing node is associated with an intramodule port and an intermodule port. The processing modules are connected in a ring via intermodule connections between at least a portion of the intermodule ports of the processing modules. A network switch is arranged in a center of the ring of processing modules and connections are formed between the network switch and at least one of the processing modules by connecting every Sth processing module to the network switch, connecting every Sth and Sth−1 processing modules to the network switch, or by connecting every Sth and Sth−r processing modules to the network switch. S is a number of steps between the 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 intramodule port and an intermodule port;
a ring formed of the processing modules via intermodule connections between at least a portion of the intermodule ports of the processing modules;
a network switch arranged in a center of the ring of processing modules;
connections formed between the network switch and at least one of the processing modules, comprising at least one of:
a connection between the network switch and only every Sth processing module, wherein S is a number of steps between the processing modules and is greater than 1;
a connection between the network switch and only every Sth and Sth−1 processing modules, wherein S is greater than 2; and
a connection between the network switch and only every Sth and Sth−r processing modules, wherein r is a different number of steps between the processing modules than the number of steps of S, which is greater than 2.
2. A computing system framework according to claim 1 , wherein the intermodule connection between each pair of the Sth and the Sth−1 processing modules is removed.
3. A computing system framework according to claim 1 , wherein the processing module N and the processing module N+S−1 are connected to the network switch, when every Sth and Sth−1 processing modules are connected to the network switch.
4. A computing system framework according to claim 1 , wherein the processing module N and the processing module N+S−r are connected to the network switch when every Sth and Sth−r processing modules are connected to the network switch, wherein M is a number of the processing modules.
5. A computing system framework according to claim 1 , wherein the ring of processing modules is connected to a further ring of processing modules by linking via a cable one of the intermodule ports on one of the processing modules of the ring with one of the intermodule ports on one of the processing modules of the further ring.
6. A computing system framework according to claim 1 , further comprising:
one or more further network switches to connect at least one of an external device and one or more further rings of processing nodes to the ring of processing nodes.
7. A computing system framework according to claim 1 , wherein data packets are transferred through the intermodule connections from one of the processing nodes of one of the processing modules in the ring to another processing node of another processing module in the ring.
8. A computing system framework according to claim 7 , wherein traffic of the data packet transfer is measured and the intermodule connections are amended based on the traffic.
9. A computing system framework according to claim 1 , wherein at least one of the intermodule ports on one of the processing modules is connected to at least one intermodule port on another processing module using a hybrid system.
10. 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 intramodule port and an intermodule port;
connecting the processing modules in a ring via intermodule connections between at least a portion of the intermodule ports of the processing modules;
arranging a network switch in a center of the ring of processing modules;
forming connections between the network switch and at least one of the processing modules, comprising at least one of:
connecting only every Sth processing module to the network switch, wherein S is a number of steps between the processing modules and is greater than 1;
connecting only every Sth and Sth−1 processing modules to the network switch, wherein S is greater than 2; and
connecting only every Sth and Sth−r processing modules to the network switch, wherein r is a different number of steps between the processing modules than the number of steps of S, which is greater than 2.
11. A method according to claim 10 , wherein the intermodule connection between each pair of the Sth and the Sth−1 processing modules is removed.
12. A method according to claim 10 , wherein the processing module N and the processing module N+S−1 are connected to the network switch, when every Sth and Sth−1 processing modules are connected to the network switch.
13. A method according to claim 10 , wherein the processing module N and the processing module N+S-r are connected to the network switch when every Sth and Sth−r processing modules are connected to the network switch, wherein M is a number of the processing modules.
14. A method according to claim 10 , further comprising:
connecting to the ring of processing modules a further ring of processing modules, wherein each ring of processing nodes comprises a block, by linking via a cable one of the intermodule ports on one of the processing modules of one block with one of the intermodule ports on one of the processing modules of the other block.
15. A method according to claim 10 , further comprising:
connecting to the ring of processing nodes at least one of an external device and one or more further rings of processing nodes via one or more further network switches.
16. A method according to claim 10 , wherein data packets are transferred through the intermodule connections from one of the processing nodes of one of the processing modules in the ring to another processing node of another processing module in the ring.
17. A method according to claim 16 , further comprising:
measuring traffic of the data packet transfer; and
amending the intermodule connections based on the traffic.
18. A method according to claim 10 , wherein at least one of the intermodule ports on one of the processing modules is connected to at least one intermodule port on another processing module using a hybrid system.