Packet spraying for load balancing across multiple packet processors

摘要

A network device includes multiple packet processing engines implemented in parallel with one another. A spraying component distributes incoming packets to the packet processing engines using a spraying technique that load balances the packet processing engines. In particular, the spraying component distributes the incoming packets based on queue lengths associated with the packet processing engines and based on a random component. In one implementation, the random component is a random selection from all the candidate processing engines. In another implementation, the random component is a weighted random selection in which the weights are inversely proportional to the queue lengths.

主权项

1. A method comprising:
determining, by a device, a first amount of space being used in a first buffer of a plurality of buffers; determining, by the device, a first weight, for the first buffer, based on a difference between a threshold and the first amount of space being used in the first buffer; determining, by the device, a second amount of space being used in a second buffer of the plurality of buffers; determining, by the device, a second weight, for the second buffer, base on a difference between the threshold and the second amount of space being used in the second buffer; selecting, by the device, the first buffer based on the first weight; selecting, by the device, the second buffer based on the first weight and the second weight; and transmitting, by the device, a particular packet to the second buffer after selecting the second buffer.