amqprs has much better performance than lapin: basic_publish benchmarking

[gftea]

Benchmark settings

• variants of message size: First 28 Fibonacci number
• total number of messages: 28 x 1024

Measure the wall-clock time
Run log:
https://github.com/gftea/amqprs/actions/runs/4142767627/jobs/7163827661

For more information about benchmarking, see Benchmark Readme