As I understand the benefit of using redis cache is that it is faster as avoid the additional hop to the database keeping important data ready to use instantly. However if Fly.io is using upstash, how can this work? Looking at the connection string that I’m provided it ends in ***.upstash.io:6379 - this is not a local address meaning that this is a network hop that would take at best 10-20ms to complete. This wouldn’t be as efficient as persisting to a database.
The docs say this:
Your databases run within Fly.io infrastructure, but not inside your organization’s network. So you’re only paying Upstash pricing - not additional VM costs.
So, at a guess, even though the address is on upstash.io, the instances provisioned for Fly customers are appearing inside Fly infra. I wonder if the subdomain is resolved differently.
Source: Upstash for Redis®* · Fly Docs
This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.