[Bug] Unexpected change in throughput running z_sub_thr and restarting z_pub_thr

[jackg0]

Describe the bug

Hi,

Thanks for all the great work on zenoh and zenoh-c! We're excited to use this project in various applications.

I am seeing varying bandwidth when running the z_sub_thr example and restarting the z_pub_thr example with a payload size of 1 MiB.

I'd note that I modified z_sub_thr to set a value of N == 100 so that stats were reported at a reasonable pace.

Using the attached annotated stdout for z_sub_thr as an example, for the first three runs of z_pub_thr 1048576 I see a nominal msg/s reported by z_sub_thr of 1300 msg/s. After running z_pub_thr for a 4th time, I see a much higher bandwidth of 4000 msg/s to 5000 msg/s.

Is this an issue with the examples, my build of zenoh-c, or possibly some other issue?

z_sub_thr_output.txt

To reproduce

1. Start the z_sub_thr subscriber. I modified this to set N == 100 as the default value for N is high for a payload of 1 MiB.
2. Start/restart the z_pub_thr publisher with a payload size of 1048576. I.e., run with z_pub_thr 1048576.

System info

• Platform: Docker container running Ubuntu 24.04
• CPU: Intel(R) Core(TM) i7-9750H CPU @ 2.60GHz
• zenoh-c commit: 6e161ac and 68ab0c1
• gcc 9.5.0 and rustc 1.75.