Live data from Hacker News

I took all my projects off the cloud, saving thousands of dollars

rameerez.com

211–220 of 406 posts

Re: I took all my projects off the cloud, saving thousands of dollars

#211
post #35

I'd be more interested to understand (from folk who were there) what the conditions were that made AWS et al such a runaway hit. What did folks gain, and have those conditions meaningfully changed in some way that makes it less of a slam dunk? My recollection from working at a tech company in the early 2010s is that renting rack space and building servers was expensive and time consuming, estimating what the right ha…

raises hand I ran a small SaaS business in the early 2000s, pre-AWS.

Renting dedicated servers was really expensive. To the extent that it was cheaper for us to buy a 1U server and host it in a local datacenter. Maintaining that was a real pain. Getting the train to London to replace a hard drive was so much fun. CDNs were "call for pricing". EC2 was a revelation when it launched. It let us expand as needed without buying servers or paying for rack space, and try experiments without shoving everything onto one server and fiddling with Apache configs in production. Lambda made things even easier (at the expense of needing new architectures).

The thing that has changed is that renting bare metal is orders of magnitude cheaper, and comparable in price to shared hosting in the bad old days.

Re: I took all my projects off the cloud, saving thousands of dollars

#212
post #16

I dislike those black and white takes a lot. It's absolutely true that most startups that just run an EC2 instance will save a lot of cash going to Hetzner, Linode, Digital Ocean or whatever. I do host at Hetzner myself and so do a lot of my clients. That being said, the cloud does have a lot of advantages: - You're getting a lot of services readily available. Need offsite backups? A few clicks. Managed database? A f…

> A few clicks. Getting through AWS documentation can be fairly time consuming.

And learning TypeScript and CDK, if we're comparing scripted, repeatable setups which you should be doing from the start.

Re: I took all my projects off the cloud, saving thousands of dollars

#213

Earlier quoted context omitted.

Hmmm, I think you have to figure out how to do your database backups anyway as trying to get a restorable backup out of RDS to use on another provider seems to be a difficult task. Backups that are stored with the same provider are good, providing the provider is reliable as a whole. (Currently going through the disaster recovery exercise of, "What if AWS decided they didn't like us and nuked our account from orbit."…

aws would never do that :) plus you can also do it in aws with like 75 clicks around UI which makes no sense even when you are tripping on acid

> 75 clicks

Well 2 commands...

  aws rds export-task create \
    --source-arn  \
    --s3-bucket-name  \
    --iam-role-arn 

Then copy it down

  aws s3 cp \
     \
     --recursive
The biggest effort would be then running the Apache Parquet to CSV tool on it.

Re: I took all my projects off the cloud, saving thousands of dollars

#214

Earlier quoted context omitted.

On this site, I've seen these kind of takes repeatedly over the past years, so I went ahead and built a little forum that consists of a single Rust binary and SQLite. The binary runs on a Mac Mini in my bedroom with Cloudflare tunnels. I get continuous backups with Litestream, and testing backups is as trivial as running `litestream restore` on my development machine and then running the binary. Despite some pages is…

Just one of the couple dozen databases we run for our product in the dev environment alone is over 12 TB. How could I not use the cloud?

what are you doing that you have 12TB in dev??? my startup isn't even using a TB in production and we hands multiple millions of dollars in transactions every month.

Re: I took all my projects off the cloud, saving thousands of dollars

#215
Up to a point, cloud can actually be very cost effective. If you are a small company, you don't actually need a lot. I ran our current service for almost one and a half year on Google Cloud Run for free. It stayed below the freemium threshold. Eventually I transitioned it over to two simple VMs. Main reason for this was that I abused these servers to also do background processing and Cloud Run is not very suitable for that as it can shut down servers in between requests and/or starve them of resources. But a small vm only costs something like 20$/month. Add another 25 or so for the load balancer and a Redis, and a few other bits and we're looking at ~100$ per month. That's very reasonable. That's not a cost worth optimizing without a very good reason. A good devops freelance person asks more per hour than that.

Where a lot of companies go wrong is premature scaling and overengineering. As you might guess from the above, I don't run kubernetes. Reason, the smallest cluster I can manage would probably be more expensive than the entirety of my setup. Before I even deploy anything. And it might not be big enough even.

I also don't run a managed postgresql. I'd like to. But I can't justify dropping north of 300$/month on that currently. A single node vm with backups is good enough. For now. I might go active master - passive master. Do I need a 3 or more node cluster? Not right now. I might one day. I always have the option to upgrade. But until then my DB is going to not break the bank. So, I run with a more modest/risky setup. It's a managed risk. I can buy it off when that becomes relevant.

The other day, our CEO vibe coded a cost calculator thingy. He did a nice job. It had to be deployed somewhere. So, I used Cloud Run. Running a small thing with essentially no traffic is as close to free as you can get. I slapped some basic authentication on it and used a git hub action that we run from the freemium allocated build minutes each month. I reviewed our Github bills a few days ago over the years: The most we payed was a few months where we were running out of build minutes in 2024. I payed 55cents for one of those bills. Our setup is cloud based but frugal. I could probably lower our bills by double digit percentages by moving everything into Hetzner. But it's literally not worth my time.

Cloud on the cheap is just really good value. You can host low traffic static websites on Cloudflare. If you have a load balancer in Google cloud (25$/month or so) you can use buckets and their CDN for next to nothing as an alternative. A small Redis costs next to nothing. We run a slightly expensive vm for batch jobs for a maybe ~2 hours per month in total. Each job only takes a few minutes. The machine is suspended in between jobs and we only pay for the hours it runs. Next to nothing.

So, yes, you can get some bare metal in Hetzner or wherever for next to nothing. But then you need to manage that setup and you need to put the time in to do that. Your time even if you don't value it personally is just about the most scarce and expensive resource a startup has if you are the CTO. The difference between you doing valuable work that moves the company forward or you procrastinating and penny pinching over stuff that does not matter can literally make or break the company.

If you spend 5K on infrastructure per month, that's a the same as a very low salary for a full time person. If you are spending that much, definitely do something because it's an obscene amount. Especially if you are early stage and have no traffic. But if you need to put somebody on that that costs closer to 10K/month for a few weeks or months, that's a project that has a pretty high price tag as well. So cutting 5K down to 1K is worth doing. But spending 10K to cut 1K to 500/month is probably going to take you years to earn back. And you lose time doing it.

Our startup is bootstrapped. We run on a bit of revenue now and our savings for most of the past five years. Anything I spend on stuff that isn't needed literally is money that otherwise would go to my bank account (via a salary raise or bonus). I understand frugality. A lot of this stuff is people getting spoiled with investor cash and spending like there is no tomorrow. If you have no accountability you optimize the infrastructure for how good it makes your resume look, not for cost.

Re: I took all my projects off the cloud, saving thousands of dollars

#216
post #89

I'm fully aware this is pedantic, but you can't save 10x. You can pay 1/10. You can save 90%. Your previous costs could have been 10x your current costs. But 10x is more by definition, not less. You can't save it.

In English, x or time(s) after a number marks a "unit" used by various verbs. A 10x increase. Increase by 10x. Go up 10x. Some of these verbs are negative like decrease or save. "Save 10x" is the same as "divide by 10". Four times less, 5 times smaller etc. are long attested.

Thats not how English works.

Saved 10x would imply there was an amount being saved that they multipled.

Re: I took all my projects off the cloud, saving thousands of dollars

#217
post #16

I dislike those black and white takes a lot. It's absolutely true that most startups that just run an EC2 instance will save a lot of cash going to Hetzner, Linode, Digital Ocean or whatever. I do host at Hetzner myself and so do a lot of my clients. That being said, the cloud does have a lot of advantages: - You're getting a lot of services readily available. Need offsite backups? A few clicks. Managed database? A f…

To me DO is a cloud. It is pricey (for performance) and convenient. It is possibly a wiser bet than AWS for a startup that wants to spend less developer (read expensive!) time on infra.

Re: I took all my projects off the cloud, saving thousands of dollars

#218

What if the data center that has your dedicated server burns down? You need to prepare to that buy renting another server at a different data center and synching them. Not a big problem, but needs to be mentioned

Correct AWS gives you availability zones. It is easy to spin up a cluster over them and have little or no downtime.

But a burning data centre is just one problem. Another is e.g. the USE1 issue AWS had. AZ didnt help here. Having compute in a other region an rerouting might. But maybe we aint talking startup concerns here. (If you have a simple enough "app" just deploy again in another region and update dns.)

Re: I took all my projects off the cloud, saving thousands of dollars

#219
I never understood the appeal of AWS like cloud for non corporate. Sure, when you run a BigCo, you have all the money and people to throw at big cloud providers, obtaining pointless certificates, and configuring obscure IAM roles.

But why small businesses like them, I can’t I understand. I host all my stuff on Hetzner. It’s easy, it’s affordable and I “move fast”. I don’t need certificates to be able to manage the servers, and there is tone of information available online. But maybe it’s because I’m old, and I used to ssh to a Linux machine, git pull the latest version, and call in a “deploy”. But modern developed seems to be afraid of raw machines.

The only thing I miss is a good transactional email provider with PAYG pricing, and EU data residency.

Re: I took all my projects off the cloud, saving thousands of dollars

#220

Cheap shot maybe, but the fact that the page takes 10 seconds to load when it hits the HN front page is a great, inadvertant illustration of why you might want to use the cloud sometimes.

You want to architect. Thats it. You can handle the 50krps that you wont get from HN from a dedicated box.
Post reply on HN