Is the 2500ms Runtime for My API Endpoint Too Long, and Can I Improve it Without Code Optimization While Using Pipedream's Free Tier?

This topic was automatically generated from Slack. You can find the original thread here.

I’ve made an api endpoint that returns a payload after collecting data from other apis. It takes about 2500ms to run. In the meantime the user is waiting. Is this good or bad, and is there anything I can do (other then optimizing the code)? I’m currently on the free tier of Pipedream, if that makes any difference (btw! I’m a newbie - this is my first time working with apis and creating endpoint)

Hi , for your case I would suggest you to measure the response time of your external API call first. So for example, if the external API response in 2000ms, then there’s not much we can do on Pipedream.

On Pipedream side, there might be a delay because of Cold Start. You can read more about it and how to eliminate it here

Just to add on, we have also recently released Warm Workers for addressing this cold start problem without any custom configuration on your side:

Hey, and @U03CXTHK4RF! Thanks for your feedback. Tried out the cold start now. It seems that it uses 6-7 sec starting from cold, but 1.5 sec when the workflow has been running. Added some logging to the code, and it looks like its one of the other API requests that takes a long time. Again, thanks for your help!