New blog post:
-
Zach Leatherman :11ty:replied to Stuart Langridge on last edited by
@sil ha! incredible parallel
-
benschwarzreplied to Zach Leatherman :11ty: on last edited by
@zachleat if this were Twitter Vercel staff would be all up in replies explaining what happened… but they aren’t, they never left Twitter.
To me the fact they Vercel and all thier staff were able to look past what X has become for the sake of their hype machine is a bad signal
-
Zach Leatherman :11ty:replied to benschwarz on last edited by
@benschwarz …yeah. It isn’t just Vercel—a lot of the JS-dev ecosystem is that way. Definitely hurt my career when everything split.
-
benschwarzreplied to Zach Leatherman :11ty: on last edited by
@zachleat I can use a boatload more 🤯 emoji if that helps?
-
Zach Leatherman :11ty:replied to benschwarz on last edited by
@benschwarz the only engagement that performs a meaningful function on mastodon is boost (favorites are for good feelings only)
-
Francis 🏴☠️ Gulottareplied to Zach Leatherman :11ty: on last edited by
> Mind-boggling that the same service would use ~1276 GB-Hrs per month on Vercel and ~101 GB-Hrs on AWS. Mind-boggling that the same service would use ~1276 GB-Hrs per month on Vercel and ~101 GB-Hrs on AWS.
This is with all the caching in place, not the same workload hitting lambda, right?
In either case yes mind boggling, and I used to run the team that owned that.
-
Zach Leatherman :11ty:replied to Francis 🏴☠️ Gulotta on last edited by
@reconbot yeah, with caching—71% cache misses on the Vercel side. ~85% cache *hits* from CloudFront.
-
Zach Leatherman :11ty:replied to Zach Leatherman :11ty: on last edited by
Found some stats:
28.8% cache hits on Vercel
81.2% cache hits on CloudFront -
Ryan Townsendreplied to Zach Leatherman :11ty: on last edited by
@zachleat all CDNs are not created equal when it comes to cache eviction.
Yet another thing often selected because of some flashy feature or promise that has knock-on significant impact on origin load and financial implications
Remember folks: the difference between 95% and 90% cache hit ratio is a doubling of origin traffic!
-
Zach Leatherman :11ty:replied to Ryan Townsend on last edited by
@ryantownsend what‘s the difference between 90% and 28%
-
Ryan Townsendreplied to Zach Leatherman :11ty: on last edited by
@zachleat bankruptcy.
-
evanareplied to Zach Leatherman :11ty: on last edited by
@zachleat I'm curious if you looked at Google's serverless products. In particular, Cloud Run also has a free tier of a similar size to Lambda, but it makes it easy to run a Docker container of the size you want without having to do the layers thing yourself.
I think the UX is also better, though they expect that you just have a container that answers HTTP, and don't put an API gateway in front of it (no API gateway charges, though).
-
Zach Leatherman :11ty:replied to evana on last edited by
@evana trying to avoid using docker, for whatever that’s worth
-
evanareplied to Zach Leatherman :11ty: on last edited by
@zachleat that's fine. I just figured that the packaging might be simpler than some of the other layer options, but I'm already in that ecosystem.
-
Zach Leatherman :11ty:replied to evana on last edited by
@evana yeah, that’s fair! I’ll keep it in the brain bank as a possible alternative!
I secretly hope someone writes a load balancer so that I can use all of the free tiers simultaneously
-
Zach Leatherman :11ty:replied to Zach Leatherman :11ty: last edited by [email protected]
Update after one billable month: the 11ty Screenshots service cost $1.95 on AWS
1.16M screenshots served
86.2% cache hits -
Bob Monsourreplied to Zach Leatherman :11ty: last edited by
@zachleat With this edit, you answered my question.
-
Andy Bellreplied to Zach Leatherman :11ty: last edited by
@zachleat cost them more to raise an invoice for you lol
-
Zach Leatherman :11ty:replied to Andy Bell last edited by
@andy this could all be solved with an even more generous free tier
-
Erik Kroes ✅replied to Zach Leatherman :11ty: last edited by
@zachleat Do you have a sweet sponsor for that?