Systems and methods for load balancing multicast traffic

Patent metadata

Field	Value
Publication number	US-9407555-B2
Application number	US-201414231225-A
Country	US
Kind code	B2
Filing date	Mar 31, 2014
Priority date	Mar 31, 2014
Publication date	Aug 2, 2016
Grant date	Aug 2, 2016

Abstract

A computer-implemented method for load balancing multicast traffic may include (1) identifying a plurality of switches that include at least a first switch that is connected to a second switch by a first path and a second path, (2) calculating a plurality of multicast distribution trees for distributing multicast traffic among the plurality of switches that includes (i) a first tree that includes the first path and whose root is different than the root of a second tree and (ii) the second tree that includes the second path, (3) receiving a plurality of multicast packets ingress to the plurality of switches at the first switch, and (4) using at least two of the plurality of multicast distribution trees to transmit the plurality of multicast packets from the first switch to the second switch. Various other methods, systems, and computer-readable media are also disclosed.

First claim

What is claimed is:

1. A computer-implemented method comprising:
identifying a plurality of switches that comprise at least a first switch and a second switch, wherein:
the first switch is connected to the second switch by a first path that extends from the first switch to the second switch;
the first switch is also connected to the second switch by a second path that extends from the first switch to the second switch;

calculating a plurality of multicast distribution trees for distributing multicast traffic among the plurality of switches by:
selecting a first root switch from the plurality of switches;
selecting a second root switch from the plurality of switches;
generating a first tree for distributing multicast traffic among the plurality of switches that is rooted on the first root switch and that includes the first path;
generating a second tree for distributing multicast traffic among the plurality of switches that is rooted on the second root switch and that includes the second path, wherein
the first root switch of the first tree is different than the second root switch of the second tree;

receiving a plurality of multicast packets ingress to the plurality of switches at the first switch; and
using, at each switch along the first path, the first tree to transmit a first portion of the plurality of multicast packets from the first switch to the second switch via the first path;
using, at each switch along the second path, the second tree to transmit a second portion of the plurality of multicast packets from the first switch to the second switch via the second path.

2. The method of claim 1 , wherein:
the plurality of switches comprise a virtual-chassis fabric that behaves as a single logical switch;
using the first tree and the second tree comprises load balancing the plurality of multicast packets across the first path and the second path.

3. The method of claim 1 , wherein calculating the plurality of multicast distribution trees comprises calculating, for each switch in the plurality of switches, a multicast distribution tree rooted on the switch.

4. The method of claim 1 , wherein calculating the plurality of multicast distribution trees comprises:
selecting, for each of the multicast distribution trees, one of the plurality of switches to be a root of the multicast distribution tree; and
calculating the multicast distribution tree such that the selected switch is the root of the multicast distribution tree.

5. The method of claim 4 , wherein the root of the multicast distribution tree is selected based at least in part on a physical topology of the plurality of switches.

6. The method of claim 4 , wherein the root of the multicast distribution tree is selected based at least in part on the root of the multicast distribution tree being a hub.

7. The method of claim 4 , wherein the root of the multicast distribution tree is selected based at least in part on input from an administrator of the plurality of switches.

8. The method of claim 4 , wherein the multicast distribution tree is calculated by the root of the multicast distribution tree.

9. The method of claim 4 , wherein the multicast distribution tree is calculated by one of the plurality of switches that is not the root of the multicast distribution tree.

10. The method of claim 1 , wherein each of the plurality of distribution trees comprises a path from the root of the distribution tree to each of the plurality of switches.

11. The method of claim 1 , wherein:
calculating the plurality of multicast distribution trees comprises calculating, for each switch in the plurality of switches, a multicast distribution tree rooted on the switch; and
using the first tree and the second tree comprises using each of the plurality of multicast distribution trees to transmit the plurality of multicast packets from the first switch to the second switch.

12. The method of claim 1 , wherein the plurality of multicast packets are of a single multicast group.

13. The method of claim 12 , wherein the multicast group comprises a virtual area network.

14. The method of claim 12 , wherein the multicast group comprises an internet protocol multicast group.

15. The method of claim 1 , wherein:
the plurality of multicast packets comprise:
a plurality of multicast packets of a first multicast group; and
a plurality of multicast packets of a second multicast group; and

using the first tree and the second tree comprises:
using the first tree rather than the second tree to transmit the plurality of multicast packets of the first multicast group from the first switch to the second switch; and
using the second tree rather than the first tree to transmit the plurality of multicast packets of the second multicast group from the first switch to the second switch.

16. The method of claim 15 , further comprising:
receiving an additional plurality of multicast packets of the first multicast group and an additional plurality of multicast packets of the second multicast group ingress to the plurality of switches at a third switch in the plurality of switches;
using the first tree rather than the second tree to transmit the additional plurality of multicast packets of the first multicast group from the third switch to the second switch; and
using the second tree rather than the first tree to transmit the additional plurality of multicast packets of the second multicast group from the third switch to the second switch.

17. A system comprising:
a memory that stores:
an identifying module that identifies a plurality of switches that comprise at least a first switch and a second switch, wherein:
the first switch is connected to the second switch by a first path that extends from the first switch to the second switch; the first switch is also connected to the second switch by ama˜l-a second path that extends from the first switch to the second switch;
a calculating module that calculates a plurality of multicast distribution trees for distributing multicast traffic among the plurality of switches by:
selecting a first root switch from the plurality of switches;
selecting a second root switch from the plurality of switches;
generating a first tree for distributing multicast traffic among the plurality of switches that is rooted on the first root switch and that includes the first path;
generating a second tree for distributing multicast traffic among the plurality of switches that is rooted on the second root switch and that includes the second path, wherein
a receiving module that receives a plurality of multicast packets ingress to the plurality of switches at the first switch; a transmitting module that: uses, at each switch along the first path, the first tree to transmit a first portion of the plurality of multicast packets from the first switch to the second switch via the first path;
uses, at each switch along the second path, the second tree to transmit a second portion of the plurality of multicast packets from the first switch to the second switch via the second path; and
at least one physical processor coupled to the memory that executes the identifying module, the calculating module, the receiving module, and the transmitting module.

18. The system of claim 17 , wherein:
the plurality of switches comprise a virtual-chassis fabric that behaves as a single logical switch
the transmitting module uses the first tree and the second tree to load balance the plurality of multicast packets across the first path and the second path.

19. The system

Assignees

• Juniper Networks Inc

Inventors

• Wu Qiang
• Hong Xichun

Classifications

• H04L45/24

  Multipath · CPC title

• H04L49/70

  Virtual switches · CPC title

• H04L15/16

  with keyboard co-operating with code discs · CPC title

• H04L45/66

  Layer 2 routing, e.g. in Ethernet based MAN's · CPC title

• H04L12/18

  for broadcast or conference {, e.g. multicast} · CPC title