Generating EGFT network topologies

• US 9,385,922 B2
• Filed: 10/31/2011
• Issued: 07/05/2016
• Est. Priority Date: 10/31/2011
• Status: Active Grant

• Current Assignee: Hewlett Packard Enterprise Development LP (Hewlett-Packard Enterprise Company)
• Original Assignee: Hewlett Packard Enterprise Development LP (Hewlett-Packard Enterprise Company)
• Inventors: Mudigonda, Jayaram, Yalagandula, Praveen, Mogul, Jeffrey Clifford
• Primary Examiner(s): Pezzlo, John
• Application Number:
  US13/285,803
• Publication Number:
  US 20130107759A1
• Time in Patent Office: 1,709 Days
• Field of Search: 370254-256, 370/252, 370/386
• US Class Current: 1/1
• CPC Class Codes: H04L 41/12 Discovery or management of network topologies
• CPC Class Codes: H04L 45/245 Link aggregation, e.g. trunking
• CPC Class Codes: H04L 45/48 Routing tree calculation
• CPC Class Codes: Y02D 30/50 in wire-line communication networks, e.g. low power modes or reduced link rate

Abstract

A method of generating a plurality of potential generalized fat tree network (GFT) topologies is provided herein. The method includes receiving parameters that specify a number of servers, a number of switches, and a number of ports in the switches. The parameters are for configuring an GFT topology. The method also includes generating a plurality of potential GFT topologies based on heuristics that favor higher bandwidth and lower latency. Generating the plurality of potential GFT topologies includes generating each level of the potential GFT topologies, recursively, and from a bottom level of a potential GFT topology to a top level.

Claims

1. A method performed by a computer processor, the method comprising:

receiving parameters that specify a number of servers, a number of switches, and a number of ports in the switches, for configuring an extended generalized fat tree network (EGFT) topology; and

generating a plurality of potential EGFT topologies based on the parameters and based on heuristics that favor higher bandwidth and lower latency, wherein generating the plurality of potential EGFT topologies comprises generating each level of the potential EGFT topologies recursively and from a bottom level of a potential EGFT topology to a top level.

2. The method recited in claim 1, comprising configuring an EGFT network topology based on one of the potential EGFT topologies.

3. The method recited in claim 1, wherein the heuristics comprise generating the top level with a maximum link aggregation (LAG) factor.

4. The method recited in claim 1, wherein the heuristics comprise generating levels of the potential EGFTs that achieve an oversubscription ratio equal to or greater than that of levels below the generated levels.

5. The method recited in claim 1, wherein the heuristics comprise assigning a current level of a potential EGFT to the top level if all low level modules can be combined into a single tree.

6. The method recited in claim 1, wherein the heuristics comprise generating a top level wherein a number of switches specified is limited to an amount that covers all uplinks from a next lower level.

7. The method recited in claim 1, wherein the parameters comprise a bisection bandwidth threshold, wherein the potential EGFTs comprise a bisection bandwidth greater than or equal to the bisection bandwidth threshold.

8. The method recited in claim 1, wherein the potential EGFTs comprise:

a plurality of parameterized levels;

a plurality of link aggregation factors at each of the parameterized levels;

a number of modules aggregated at each of the parameterized levels; and

a number of top switches aggregated at each of the parameterized levels.

9. A computer system for generating a plurality of potential extended generalized fat trees, comprising:

a memory storing instructions;

a processor configured to execute the instructions to;

receive parameters that specify a number of servers, a number of switches, and a number of ports in the switches, for configuring an extended generalized fat tree network (EGFT) topology; and

generate a plurality of potential EGFT topologies based on the parameters, and based on heuristics that favor higher bandwidth and lower latency, wherein generating the plurality of potential EGFT topologies comprises by generating each level of the potential EGFT topologies recursively and from a bottom level of a potential EGFT topology to a top level, wherein the potential EGFTs comprise;

a plurality of parameterized levels;

a plurality of link aggregation factors at each of the parameterized levels; and

a number of top switches aggregated at each of the parameterized levels.

10. The computer system recited in claim 9, wherein the heuristics comprise generating the top level with a maximum link aggregation (LAG) factor.

11. The computer system recited in claim 9, wherein the heuristics comprise generating levels of the potential EGFTs that achieve an oversubscription ratio equal to or greater than that of levels below the generated levels.

12. The computer system recited in claim 9, wherein the heuristics comprise assigning a current level of a potential EGFT to the top level if all low level modules can be combined into a single tree.

13. The computer system recited in claim 9, wherein the heuristics comprise generating a top level wherein a number of switches specified is limited to an amount that covers all uplinks from a next lower level.

14. The computer system recited in claim 9, wherein the parameters comprise a bisection bandwidth threshold, wherein the potential EGFTs comprise a bisection bandwidth greater than or equal to the bisection bandwidth threshold.

15. A non-transitory, computer-readable medium comprising machine-readable instructions executable by a processor to:

receive parameters that specify a number of servers, a number of switches, and a number of ports in the switches, for configuring an extended generalized fat tree network (EGFT) topology; and

generate a plurality of potential EGFT topologies based on the parameters, and based on heuristics that favor higher bandwidth and lower latency, wherein generating the plurality of potential EGFT topologies comprises generating each level of the potential EGFT topologies recursively and from a bottom level of a potential EGFT topology to a top level, wherein the parameters comprise a bisection bandwidth threshold, and wherein the potential EGFTs comprise a bisection bandwidth greater than or equal to the bisection bandwidth threshold.

16. The non-transitory, computer-readable medium recited in claim 15, wherein the potential EGFTs comprise:

a plurality of parameterized levels;

a plurality of link aggregation factors at each of the parameterized levels;

a number of modules aggregated at each of the parameterized levels; and

a number of top switches aggregated at each of the parameterized levels.

17. The non-transitory, computer-readable medium recited in claim 16, wherein the heuristics comprise generating the top level with a maximum link aggregation (LAG) factor.

18. The non-transitory, computer-readable medium recited in claim 17, wherein the heuristics comprise generating levels of the potential EGFTs that achieve an oversubscription ratio equal to or greater than that of levels below the generated levels.

19. The non-transitory, computer-readable medium recited in claim 18, wherein the heuristics comprise assigning a current level of a potential EGFT to the top level if all low level modules can be combined into a single tree.

20. The non-transitory, computer-readable medium recited in claim 19, wherein the heuristics comprise generating a top level wherein a number of switches specified is limited to an amount that covers all uplinks from a next lower level.