I am working on a wrapper api that calls another sub-api from inside. I used “requests.get” method initially.
I moved this wrapper api to GKE and the sub-api remains on a GCP cloud VM, both on the same network and in same zone.
I have observed that while calling the wrapper api, there is a delay in sub-api response. This delay ranges from negligible amount (around 2-3 milliseconds) to larger numbers like 60-70 milliseconds.
I am not able to find a solution to this issue.
Could this be due to the use of “threading” in my wrapper api to call multiple sub-api or should i look for alternatives of “requests.get” method? Please provide your insights.
It could be network or server delays and not your code.
I would use a tool like wireshark to trace the network traffic and see what the timing of the request and response packets actually are.