Methods, systems, and devices for bandwidth steering using photonic devices

Patent metadata

Field	Value
Publication number	US-12075201-B2
Application number	US-202017775820-A
Country	US
Kind code	B2
Filing date	Nov 13, 2020
Priority date	Nov 15, 2019
Publication date	Aug 27, 2024
Grant date	Aug 27, 2024

Abstract

Disclosed herein are methods, systems, and devices for bandwidth steering. Systems may include a plurality of compute nodes configured to execute one or more applications, a plurality of first level resources communicatively coupled to the plurality of compute nodes, a plurality of second level resources communicatively coupled to the plurality of first level resources, and a plurality of third level resources communicatively coupled to the plurality of second level resources. Systems may also include a plurality of optical switch circuits communicatively coupled to the plurality of first level resources and the plurality of second level resources, wherein each of the plurality of optical switch circuits is coupled to more than one of the plurality of the first level resources and is also coupled to more than one of the plurality of the second level resources.

First claim

What is claimed is:

1. A system comprising:
a plurality of compute nodes configured to execute one or more applications;
a plurality of first level resources communicatively coupled to the plurality of compute nodes;
a plurality of second level resources communicatively coupled to the plurality of first level resources;
a plurality of third level resources communicatively coupled to the plurality of second level resources; and
a plurality of optical switch circuits communicatively coupled to the plurality of first level resources and the plurality of second level resources, wherein each of the plurality of optical switch circuits is coupled to more than one of the plurality of the first level resources and is also coupled to more than one of the plurality of the second level resources.

2. The system of claim 1 , wherein the plurality of first level resources comprises a plurality of Top of Rack (ToR) switches, wherein the plurality of second level resources comprises a plurality of aggregation switches, and wherein the plurality of third level resources comprises a plurality of core switches.

3. The system of claim 2 , wherein the plurality of ToR switches and the plurality of aggregation switches is arranged in a plurality of pods.

4. The system of claim 3 , wherein each of the plurality of pods includes one of the plurality of optical switch circuits.

5. The system of claim 4 , wherein each of the plurality of optical switch circuits is communicatively coupled to one of the plurality of first level resources included in each pod, and wherein each of the plurality of optical switch circuits is communicatively coupled to one of the plurality of second level resources included in each pod.

6. The system of claim 1 , wherein the plurality of optical switch circuits are configurable to implement a designated path based on a specified network topology.

7. The system of claim 6 , wherein the specified network topology does not include any of the plurality of third level resources.

8. The system of claim 6 , wherein the designated path is determined based, at least in part, on network traffic congestion.

9. The system of claim 1 , wherein the plurality of compute nodes, the plurality of first level resources, the plurality of second level resources, the plurality of third level resources, and the plurality of optical switch circuits are included in a data center.

10. A method comprising:
identifying a network traffic event;
identifying a designated path associated with a specified network topology based, at least in part, on one or more compute nodes associated with the network traffic event; and
configuring at least one optical switch circuit based, at least in part, on the identified network topology, wherein the network topology comprises:
a plurality of first level resources communicatively coupled to the plurality of compute nodes;
a plurality of second level resources communicatively coupled to the plurality of first level resources; and
a plurality of third level resources communicatively coupled to the plurality of second level resources.

11. The method of claim 10 , wherein the plurality of first level resources comprises a plurality of Top of Rack (ToR) switches, wherein the plurality of second level resources comprises a plurality of aggregation switches, wherein the plurality of third level resources comprises a plurality of core switches, wherein the plurality of ToR switches and the plurality of aggregation switches is arranged in a plurality of pods, and wherein each of the plurality of pods includes one of a plurality of optical switch circuits.

12. The method of claim 11 , wherein each of the plurality of optical switch circuits is communicatively coupled to one of the plurality of first level resources included in each pod, and wherein each of the plurality of optical switch circuits is communicatively coupled to one of the plurality of second level resources included in each pod.

13. The method of claim 12 , wherein the specified network topology does not include any of the plurality of third level resources.

14. The method of claim 13 , wherein the plurality of compute nodes, the plurality of first level resources, the plurality of second level resources, the plurality of third level resources, and the plurality of optical switch circuits are included in a data center.

15. The method of claim 12 , wherein the designated path is identified based, at least in part, buffer utilization of at least some of the plurality of first level resources, the plurality of second level resources, and the plurality of third level resources.

16. A device comprising:
an optical switch circuit;
a first communications interface communicatively coupled to the optical switch circuit and a plurality of first level resources; and
a second communications interface communicatively coupled to the optical switch circuit and a plurality of second level resources, wherein the plurality of first level resources is communicatively coupled to a plurality of compute nodes, wherein the plurality of second level resources is communicatively coupled to the plurality of first level resources and a plurality of third level resources.

17. The device of claim 16 , wherein the plurality of first level resources comprises a plurality of Top of Rack (ToR) switches, wherein the plurality of second level resources comprises a plurality of aggregation switches, wherein the plurality of third level resources comprises a plurality of core switches.

18. The device of claim 17 , wherein the plurality of ToR switches and the plurality of aggregation switches is arranged in a plurality of pods, wherein the optical switch circuit is communicatively coupled to one of the plurality of first level resources included in each pod, and wherein the optical switch circuit is 10 communicatively coupled to one of the plurality of second level resources included in each pod, and wherein the optical switch circuit is configurable to implement a designated path based on a specified network topology.

Assignees

• Univ California

Inventors

• Michelogiannakis Georgios
• Shen Yiwen
• Teh Min Yee
• Shalf John
• Glick Madeleine
• Bergman Keren

Classifications

• H04L41/12

  Discovery or management of network topologies · CPC title

• H04Q11/0062 Primary

  Network aspects · CPC title

• H04L41/0806 Primary

  for initial configuration or provisioning, e.g. plug-and-play · CPC title

• H04Q2011/0086

  Network resource allocation, dimensioning or optimisation · CPC title

• H04Q2011/0073

  Provisions for forwarding or routing, e.g. lookup tables · CPC title