1) No, I don't think they are. Workers aren't very good at jobs that take a long time. 2) 30 seconds, generally. 15 minutes for cron triggers and queue consumers. See the docs 3) If the job can be split up into smaller jobs that take no more than 30 seconds each, you could use DOs and alarms to do the job step by step. Doing it like this would allow you to query the status in between. 4) It's not great. Postgress is better supported
I don't understand what you are referring to. Is this something you found in the docs? Something you read in a blog post? Something you read here on discord? Something you're seeing yourself?
is it fine if i globally declare a variable? such as a drizzle database instead of c.set("db")c.set("db") and access it through c.varc.var(this is for hono specifically, but same applies for workers)
I have not. If your worker is <1mb and you're on the paid tier it has previously been (and as far as I know still is) stored in all datacentres, so with extremely low latency.
The latency numbers are extremely dependant on what your worker is doing and where you are. Comparing the latencies we experience to workers wouldn't be meaningful.