Optimized placement design of network and infrastructure components

1. A method for optimizing a network composition and component configuration in a server environment, the method comprising computer-implemented steps of:

constructing a plurality of populated racks, each populated rack selected so as to contain an optimized size and configuration of a set of components constrained according to a set of user-selected requirements that specify a type and a quantity of one or more of the set of components, the set of components comprising one or more of a set of servers, a set of compute nodes, and a set of switches;

optimizing a configuration of the plurality of populated racks in the server environment, wherein the configuration is based on power requirements for the plurality of populated racks, cooling requirements for the plurality of populated racks, and weight requirements for the plurality of populated racks, wherein optimization of the configuration of the plurality of populated racks comprises:

determining a set of connection points between each rack from the plurality of populated racks, wherein each of the set of connection points includes an uplink to a central switch in the server environment;

configuring the set of populated racks from the plurality of populated racks in a substantially central portion of the server environment based on a prioritized number of uplinks to the central switch, whereby minimizing the cable length between the populated racks in the plurality of racks;

and

optimizing a position of a set of cable trays and corresponding junctions in the configured plurality of populated racks by reducing a number of cable tray channels associated with the cable length between populated racks of the plurality of populated racks.

2. The method of claim 1 , the optimizing a position of a set of cable trays and corresponding junctions further comprising minimizing the cable length and the cable tray channels while adhering to a set of constraints within the server environment attributable to any of the following: heating ventilation air condition (HVAC) equipment, stairwells, multiple floors, generators, uninterruptable power supply devices, and water supply components.

3. The method of claim 1 , the set of components further comprising one or more of the following: a network block-factor, a set of power distribution units, heating/ventilation/air-condition (HVAC) equipment, and a set of generators.

4. The method of claim 1 ,

wherein the set of populated racks has a relatively higher number of connection points and a relatively shorter connection distance.

5. The method of claim 1 , wherein the uplink to the central switch is connected to a Clos-type network, wherein a port map from a device to the central switch is defined by the configuration of the set of populated racks in the substantially central portion of the server environment, and wherein the computer implemented step of optimizing the configuration of the plurality of populated racks further comprises:

calculating a series of cost tradeoffs, for each Clos-network, for performing each of the following:

selecting different Clos-network design ratios from fully non-blocking to full over-subscription by varying combinations of switches and cables; and

providing a redundant Clos-network design to protect against failure of a single component.

6. The method of claim 1 , further comprising the computer implemented step of determining the substantially central portion of the server environment using a Floyd-Warshall algorithm that computes O(n 3 ), wherein n is the set of rack spaces from the plurality of populated racks.

7. The method of claim 1 , the constructing the plurality of populated racks comprising optimizing a composition and placement of a set of elements within each populated rack from the plurality of populated racks by arranging the set of elements in a dense configuration.

8. A system for optimizing a network composition and component configuration in a dense server environment, comprising:

a memory medium comprising program instructions;

a bus coupled to the memory medium; and

a processor, for executing the program instructions, coupled to a component automated placement (CAP) tool via the bus that when executing the program instructions causes the system to:

construct a plurality of populated racks, each populated rack selected so as to contain an optimized size and configuration of a set of components constrained according to a set of user-selected requirements that specify a type and a quantity of one or more of the set of components, the set of components comprising one or more of a set of servers, a set of compute nodes, and a set of switches;

optimize a configuration of the plurality of populated racks in the server environment wherein the configuration is based on power requirements for the plurality of populated racks, cooling requirements for the plurality of populated racks, and weight requirements for the plurality of populated racks, wherein optimization of the configuration of the plurality of populated racks comprises:

determining a set of connection points between each rack from the plurality of populated racks, wherein each of the set of connection points includes an uplink to a central switch in the server environment;

configuring the set of populated racks from the plurality of populated racks in a substantially central portion of the server environment based on a prioritized number of uplinks to the central switch, whereby minimizing the cable length between the populated racks in the plurality of racks;

and

optimize a position of a set of cable trays and corresponding junctions in the configured plurality of populated racks by reducing a number of cable tray channels associated with the cable length between populated racks of the plurality of populated racks.

9. The system according to claim 8 , the program instructions further causing the system to optimize a position of a set of cable trays and corresponding junctions further comprising minimizing the cable length and the cable tray channels while adhering to a set of constraints within the datacenter attributable to any of the following: heating ventilation air condition (HVAC) equipment, stairwells, multiple floors, generators, uninterruptable power supply devices, and water supply components.

10. The system according to claim 8 , the set of components further comprising one or more of the following: a network block-factor, a set of power distribution units, heating/ventilation/air-condition (HVAC) equipment, and a set of generators.

11. The system according to claim 8 ,

wherein the set of populated racks has a relatively higher number of connection points and a relatively shorter connection distance.

12. The system according to claim 8 , wherein the uplink to the central switch is connected to a Clos-type network, wherein a port map from a device to the central switch is defined by the configuration of the set of populated racks in the substantially central portion of the server environment, and wherein the program instructions for configuring the plurality of populated racks further causes the system to:

calculate a series of cost tradeoffs, for each Clos-network, for performing each of the following:

selecting different Clos-network design ratios from fully non-blocking to full over-subscription by varying combinations of switches and cables; and

providing a redundant Clos-network design to protect against failure of a single component.

13. The system according to claim 8 , further comprising program instructions causing the system to determine the substantially central portion of the server environment using a Floyd-Warshall algorithm that computes O(n 3 ), wherein n is the set of rack spaces from the plurality of populated racks.

14. The system according to claim 8 , the program instructions for constructing the plurality of populated racks further causing the system to optimize a composition and placement of a set of elements within each populated rack from the plurality of populated racks by arranging the set of elements in a dense configuration.

15. A computer program product for optimizing a network composition and component configuration in a server environment, the computer program product comprising a computer readable storage device, and program instructions stored on the computer readable storage device, to:

construct a plurality of populated racks, each populated rack selected so as to contain an optimized size and configuration of a set of components constrained according to a set of user-selected requirements that specify a type and a quantity of one or more of the set of components, the set of components comprising one or more of the following: a set of servers, a set of compute nodes, and a set of switches;

optimize a configuration of the plurality of populated racks in the server environment wherein the configuration is based on power requirements for the plurality of populated racks, cooling requirements for the plurality of populated racks, and weight requirements for the plurality of populated racks, wherein optimization of the configuration of the plurality of populated racks comprises:

determining a set of connection points between each rack from the plurality of populated racks, wherein each of the set of connection points includes an uplink to a central switch in the server environment;

configuring the set of populated racks from the plurality of populated racks in a substantially central portion of the server environment based on a prioritized number of uplinks to the central switch, whereby minimizing the cable length between the populated racks in the plurality of racks;

and

optimize a position of a set of cable trays and corresponding junctions in the configured plurality of populated racks by reducing a number of cable tray channels associated with the cable length between populated racks of the plurality of populated racks.

16. The computer program product of claim 15 , the computer readable storage device further comprising program instructions to optimize a position of a set of cable trays and corresponding junctions further comprising minimizing the cable length and the cable tray channels while adhering to a set of constraints within the datacenter attributable to any of the following: heating ventilation air condition (HVAC) equipment, stairwells, multiple floors, generators, uninterruptable power supply devices, and water supply components.

17. The computer program product of claim 15 , the set of components further comprising one or more of the following: a network block-factor, a set of power distribution units, heating/ventilation/air-condition (HVAC) equipment, and a set of generators.

18. The computer program product according to claim 15 ,

wherein the set of populated racks having has a relatively higher number of connection points and a relatively shorter connection distance.

19. The computer program product according to claim 15 , the computer readable storage device further comprising program instructions to determine the substantially central portion of the server environment using a Floyd-Warshall algorithm that computes O(n 3 ), wherein n is the set of rack spaces from the plurality of populated racks.

20. The computer program product according to claim 15 , the computer readable storage device further comprising program instructions to optimize a composition and placement of a set of elements within each populated rack from the plurality of populated racks by arranging the set of elements in a dense configuration.