Inquiry Regarding Zenoh Latency Measurement

[KOOBK]

Dear Zenoh Team,

I am currently evaluating Zenoh and have developed a keen interest in its performance, particularly in terms of latency. Upon reviewing the example code provided, specifically the z_ping Code, I noticed the inclusion of a Warmup Code.

When conducting tests, I observed that the latency measurements were significantly higher when the Warmup Code was removed. Similarly, even after applying sleep, the latency remained high without the Warmup Code.

In light of these observations, I would like to ask a few questions:

1. Is a warmup period essential for measuring raw latency in Zenoh? If so, could you please explain the reasoning behind this?

2. I noticed an increase in latency immediately after applying sleep, despite it not being the initial connection. Could you provide any insight into why this might be the case?

3. Are there any plans to improve latency measurements without the need for a Warmup Code?

I would greatly appreciate any detailed responses you could provide.

Thank you for your time and consideration.

Best regards,