I have some API calls to an external provider in my app. Each one takes about 10 seconds (it’s a web scraper that gets images from a website).
Bubble seems to be running these in sequence, not parallel. That means it takes like
40 seconds to run all 4 instead of 10. None of them reference data from each other, only the Current workflow’s Thing (it’s a backend trigger). The final step 12 saves the four generated images to the Thing.
I’ve always seen Bubble running API calls in parallel where possible. Any smart ideas for how to speed this up (aside from splitting it into separate scheduled workflows that just have one API call in each)? The same behaviour occurs even when step 12 is removed.
What’s the initial piece of data you have to work with? I’d assume the URL.
If so can you split this to have an initial backend workflow that just starts the process for each one?
Run on list of texts
Wf1: Scrape_scheduler
Schedules scraper for current date and time on url
Wf2: scraper
12 step flow to scrape
URL from Thing now (backend trigger). Same URL for all requests.
Yeah, but that’s what I was trying to avoid, as it means instead of 4 API calls + 1 Make changes to Thing action, I have 4 API calls, 4 Make changes to Thing, and 4 ‘add item to scheduler’ WU penalties 
Just confused why Bubble is running them all in sequence when the only dynamic data used in all of them is Thing now’s URL and an option set for the element selector.
1 Like
Is the scraper self created or using an API? Can you rework it to pass a list of URLs then have a retrieval call when you get a webhook response?
Random API. Might just have to bite the bullet and split the workflow up.
