Per-packet load-balancing over multiple network paths
Techniques are presented herein for load-balancing a sequence of packets over multiple network paths on a per-packet basis. In one example, a first network node assigns sequence numbers to a sequence of packets and load-balances the sequence of packets on a per-packet basis over multiple network paths of a network to a second network node. The second network node buffers received packets that are received over the multiple network paths of the network from the first network node and re-orders the received packets according to sequence numbers of the received packets.
1. A method comprising:
at a first network node:
assigning sequence numbers to a sequence of packets; and
load-balancing the sequence of packets on a per-packet basis over multiple network paths of a network to a second network node, including transmitting the sequence of packets in a single pseudowire over the network from the first network node to the second network node; and
at the second network node:
buffering received packets that are received over the multiple network paths of the network from the first network node; and
re-ordering the received packets according to sequence numbers of the received packets.
2. The method of claim 1 , further comprising:
at the first network node:
obtaining a sequence of bits; and
converting the sequence of bits to the sequence of packets.
3. The method of claim 2 , further comprising:
at the second network node:
converting the received packets to the sequence of bits.
4. The method of claim 1 , wherein load-balancing the sequence of packets includes:
load-balancing the sequence of packets based on respective available bandwidths of the multiple network paths.
5. The method of claim 1 , wherein load-balancing the sequence of packets includes:
load-balancing the sequence of packets based on respective packet delays of the multiple network paths.
6. The method of claim 1 , wherein the multiple network paths include respective parallel links.
7. The method of claim 1 , wherein the multiple network paths include respective network nodes.
8. The method of claim 1 , wherein load-balancing the sequence of packets includes:
communicating with the second network node over the network using private line emulation techniques or Ethernet virtual private wire service techniques.
9. The method of claim 1 , wherein the sequence of packets includes a sequence of packets that each share common flow information.
10. A system comprising:
a first network node and a second network node;
wherein the first network node is configured to:
assign sequence numbers to a sequence of packets; and
load-balance the sequence of packets on a per-packet basis over multiple network paths of a network to a second network node, including transmitting the sequence of packets in a single pseudowire over the network from the first network node to the second network node; and
wherein the second network node is configured to:
buffer received packets that are received over the multiple network paths of the network from the first network node; and
re-order the received packets according to sequence numbers of the received packets.
11. The system of claim 10 , wherein the first network node is further configured to:
obtain a sequence of bits; and
convert the sequence of bits to the sequence of packets.
12. The system of claim 11 , wherein the second network node is further configured to:
convert the received packets to the sequence of bits.
13. The system of claim 10 , wherein the first network node is further configured to:
load-balance the sequence of packets based on respective available bandwidths of the multiple network paths.
14. The system of claim 10 , wherein the first network node is further configured to:
load-balance the sequence of packets based on respective packet delays of the multiple network paths.
15. One or more non-transitory computer readable storage media encoded with instructions that, when executed by one or more processors, cause the one or more processors to:
at a first network node:
assign sequence numbers to a sequence of packets; and
load-balance the sequence of packets on a per-packet basis over multiple network paths of a network to a second network node, including transmitting the sequence of packets in a single pseudowire over the network from the first network node to the second network node; and
at the second network node:
buffer received packets that are received over the multiple network paths of the network from the first network node; and
re-order the received packets according to sequence numbers of the received packets.
16. The one or more non-transitory computer readable storage media of claim 15 , wherein the instructions further cause the one or more processors to:
at the first network node:
obtain a sequence of bits; and
convert the sequence of bits to the sequence of packets.
17. The one or more non-transitory computer readable storage media of claim 16 , wherein the instructions further cause the one or more processors to:
at the second network node:
convert the received packets to the sequence of bits.
18. The one or more non-transitory computer readable storage media of claim 15 , wherein the instructions further cause the one or more processors to:
load-balance the sequence of packets based on respective available bandwidths of the multiple network paths.
19. The one or more non-transitory computer readable storage media of claim 15 , wherein the instructions further cause the one or more processors to:
load-balance the sequence of packets based on respective packet delays of the multiple network paths.
20. The method of claim 1 , wherein the single pseudowire is configured on the first network node and the second network node.