If my math is not wrong, running a single “standard” container during 1 month (non-stop) would cost ~55$.
This looks extremely expensive for anything other than a simple demo/toy project. I can’t think of a reason I’d use this for heavy services instead of using [anything else]. Maybe I’m not seeing the use-case?
NathanFlurry 6 days ago [-]
A 1 CPU + 2 GB of RAM + 50 GB ephemeral storage on Cloudflare Containers is $74.90.
The same on Fly Machines is $31.00 (performance-1x, varies by region). Fly Machines has the same sleeping functionality as Cloudflare.
Rivet Containers also has a similar price point of $29.40, but takes a different approach to sleeping (opts for optimizing coldstarts + autoscaling over snapshotting). (I work at Rivet)
blixt 6 days ago [-]
I think Modal, which AFAIK has a similar feature set to Cloudflare Containers, also works out very favorably price wise compared to Cloudflare Containers.
rochoa 6 days ago [-]
Math is not wrong for the standard instance.
This is about using and abusing the _on-demand_ part.
The first example in the Getting started goes with sleepAfter = '10s'.
Havoc 5 days ago [-]
It’s likely aimed at bursty workloads. ie not one instance but a use case that fluctuates between 1 and 100 instances.
aiisahik 6 days ago [-]
I don't think you can calculate the cost of serverless compute this way. What containers do you have that run "non-stop"?
If the container doesn't run any workloads, it doesn't cost you anything. Most of the compute i pay for sit idle most of the time.
This is amazing pricing.
develatio 6 days ago [-]
Say I want to deploy a service that is currently receiving 1rps at a constant rate, no upticks, no gaps. Wouldn’t that be the cost? If the answer is “yes”, then no, that is a terrible pricing.
rohan_ 5 days ago [-]
Don’t host your website on containers, that’s what workers are for
psibi 21 hours ago [-]
Aren't you limited with Workers ? Like would you be able to deploy a OCaml or a Haskell application using it ?
sofixa 6 days ago [-]
Realistically, almost nobody has this type of usage. And for those that do, yes, serverless autoscaling up from zero is not appropriate.
develatio 6 days ago [-]
True. Most services (unless your service is a demo/toy project, as I stated earlier) have way more traffic. It might not be evenly distributed though the day, but if you add all the CPU time in a day, I’m sure it will exceed 24h. So that leaves me with the question for who is this? Developers that are starting and want to deploy something small?
ignoramous 2 days ago [-]
> leaves me with the question for who is this
The blog post answers this. Containers was built for folks who wanted to move rest of their workloads onto Cloudflare alongside Workers/R2/AI & other offerings.
From my experience, the Workers platform is real popular among indie developers, software shops, and shops building SaaS, who typically want zero-dev ops setup and usually pass down hosting costs to their customers.
That said, compared to new cloud providers like Fly/Railway, the pricing is indeed steep.
0xy 6 days ago [-]
And the gigantic AWS-tier bandwidth costs. This misses the mark by a lot. Classic example of pricing ruining a launch of decent technology.
It seems like always-on containers are not viable on this, so what's the point?
Egress is $25/TB unlike their other stuff. No thanks
Havoc 5 days ago [-]
Sounds similar to what bunny is offering
Think CF lets you rate limit right?some form of it seems necessary with those egress rates
Zerpiez 6 days ago [-]
Would UDP services and anycast DNS be supported in the future e.g. to be able to run dnsdist or similar services.
NathanFlurry 6 days ago [-]
They stated on the livestream they're considering TCP, but I suspect UDP is not coming soon since Workers themselves don't support UDP. All traffic going to Cloudflare Containers must be "proxied" through the Workers platform.
ttoinou 6 days ago [-]
So can we now host a whole website distributed in all regions on the edge with Cloudflare workers + containers ?
kentonv 5 days ago [-]
You've been able to do that for years, just using Workers (using Durable Objects for storage).
6 days ago [-]
Rendered at 11:42:42 GMT+0000 (Coordinated Universal Time) with Vercel.
This looks extremely expensive for anything other than a simple demo/toy project. I can’t think of a reason I’d use this for heavy services instead of using [anything else]. Maybe I’m not seeing the use-case?
The same on Fly Machines is $31.00 (performance-1x, varies by region). Fly Machines has the same sleeping functionality as Cloudflare.
Rivet Containers also has a similar price point of $29.40, but takes a different approach to sleeping (opts for optimizing coldstarts + autoscaling over snapshotting). (I work at Rivet)
This is about using and abusing the _on-demand_ part.
The first example in the Getting started goes with sleepAfter = '10s'.
If the container doesn't run any workloads, it doesn't cost you anything. Most of the compute i pay for sit idle most of the time.
This is amazing pricing.
The blog post answers this. Containers was built for folks who wanted to move rest of their workloads onto Cloudflare alongside Workers/R2/AI & other offerings.
From my experience, the Workers platform is real popular among indie developers, software shops, and shops building SaaS, who typically want zero-dev ops setup and usually pass down hosting costs to their customers.
That said, compared to new cloud providers like Fly/Railway, the pricing is indeed steep.
It seems like always-on containers are not viable on this, so what's the point?
(I work at Rivet)
Think CF lets you rate limit right?some form of it seems necessary with those egress rates