Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

We are far from a better Heroku for production apps in a hyper cloud

about.gitlab.com

Post reply on HN
Print view View original

251–260 of 275 posts

Re: We are far from a better Heroku for production apps in a hyper cloud

#251

Post by joshmanders »

cortesoft wrote: post #191
jrochkind1 wrote: post #186

Earlier quoted context omitted.

I definitely wonder why someone else hasn't managed to compete with them on price for a similar and similarly high-quality service. But this post makes me think it's unlikely to be gitlab.

Because if you charged less you wouldn’t make enough to support a business?

Disclaimer: I am building a company around this.

Not entirely. There's room for improvement on pricing. But you have to employ a culture where being cost-effective is engrained in the company.

I like to quote the pizza chain Lil Caesar's when I discuss this, which is why my company I am building is able to provide the same core features but cost less. We look for ways to save on operational costs and resources ourselves so that we can pass the buck down to our customers without compromising on quality.

Re: We are far from a better Heroku for production apps in a hyper cloud

#252

Post by sergiotapia »

jrochkind1 wrote: post #22

So I think heroku has stagnated a bit, and I am definitely very open to "better herokus". But the idea that something you did in five minutes is going to be better than heroku (in all ways?) is.... I don't know what the adjective is. Heroku is very mature and polished software, that almost always does what you expect -- it's a very non-leaky abstraction, I guess. You can almost always get away with ignoring the detai…

render.com is a better Heroku. Absurdly easy, inexpensive and killer UX.

They also have in beta a way to spin up whole environments for every PR your dev team has. Imagine every PR has it's own database, backend and frontend, spun up/killed on the fly.

It's a very compelling feature.

Re: We are far from a better Heroku for production apps in a hyper cloud

#253

Post by ZephyrBlu »

boleary-gl wrote: post #234
ZephyrBlu wrote: post #222

Earlier quoted context omitted.

A PR person did review the post. See this comment: https://news.ycombinator.com/item?id=26557589

Where does that show anyone from the PR department approving it? I don't see anyone I recognize from the PR department there... This is how "unfiltered" blogs work: https://about.gitlab.com/handbook/marketing/blog/unfiltered/

Ah sorry I misread "PR" as in Pull Request for "PR" as in Public Relations.

I believe there were at least marketing people involved in approving the post though.

Re: We are far from a better Heroku for production apps in a hyper cloud

#254

Post by brodock »

rad_gruchalski wrote: post #112
boleary-gl wrote: post #49

Earlier quoted context omitted.

I think that lots of the clouds will have their own offering, but having an open source based offering across clouds is what we're going for. Also it's very early stages - but the goal would to be better than Beanstalk

As long as it is faster than gitlab itself, sure. Every time I go to the hosted gitlab, I cringe. I like the product but the whole thing is sooo slow. First, 5 seconds wasted to "check if my browser supports gitlab", then the lag on every click. Hopefully your heroku will have a better experience than your gitlab.

Isn't that "5 seconds" a cloudflare thing?

Re: We are far from a better Heroku for production apps in a hyper cloud

#255

Post by bionhoward »

leerob wrote: post #165
danjac wrote: post #114

Earlier quoted context omitted.

Isn't Vercel pretty much just for Jamstack? Not everyone wants or needs that so I'm not sure it's better.

Hey! Lee from Vercel here. Vercel isn't just for static sites – it also handles server-rendered applications, serverless functions, and hybrid frameworks (like Next.js). The current focus has been on an incredible front-end experience, but we're working to become a first-party integrator with your favorite backend solutions. For example, you can deploy a full-stack Next.js application with a PostgreSQL database in le…

I beat this horse to death, but I would definitely use it for if there was just a way to allowlist / vpc-peer so the functions can be the only thing talking to the database. Password isn't enough for a lot of applications.

As long as vercel has lame security, I'll consider it a toy, nice for quick jokey blogs and stuff but not something to use for a serious project. Not about to risk getting sued just to use vercel simply because your stuff's easy to use

Re: We are far from a better Heroku for production apps in a hyper cloud

#256

Post by anurag »

jrochkind1 wrote: post #248
cpursley wrote: post #229

Earlier quoted context omitted.

Iny my experience, Render.com is a better (at least cheaper) Heroku plus it handles distributed Elixir really easily.

Nice, I hadn't heard of them, but does look interesting, I'll add it to the list to check out at some point. Going through the docs just a bit, I see some things have you using docker manually, like "Deploy Persistent Redis with Docker" -- that's not necessarily a barrier, but it is already less simple/abstracted than heroku. But still might be worth it. I don't know how to keep up current awareness of this stuff, wh…

(Render founder) the current Redis setup is one-click and zero maintenance, but we're also planning to launch a fully managed Redis later this year.

Re: We are far from a better Heroku for production apps in a hyper cloud

#257

Post by anurag »

cpursley wrote: post #229
jrochkind1 wrote: post #22

So I think heroku has stagnated a bit, and I am definitely very open to "better herokus". But the idea that something you did in five minutes is going to be better than heroku (in all ways?) is.... I don't know what the adjective is. Heroku is very mature and polished software, that almost always does what you expect -- it's a very non-leaky abstraction, I guess. You can almost always get away with ignoring the detai…

Iny my experience, Render.com is a better (at least cheaper) Heroku plus it handles distributed Elixir really easily.

Thanks for the plug! Here's our (admittedly biased) Render vs Heroku comparison: we're not just cheaper; we're a better fit for how applications are built today.

https://render.com/render-vs-heroku-comparison

Re: We are far from a better Heroku for production apps in a hyper cloud

#258

Post by leerob »

bionhoward wrote: post #255
leerob wrote: post #165

Earlier quoted context omitted.

Hey! Lee from Vercel here. Vercel isn't just for static sites – it also handles server-rendered applications, serverless functions, and hybrid frameworks (like Next.js). The current focus has been on an incredible front-end experience, but we're working to become a first-party integrator with your favorite backend solutions. For example, you can deploy a full-stack Next.js application with a PostgreSQL database in le…

I beat this horse to death, but I would definitely use it for if there was just a way to allowlist / vpc-peer so the functions can be the only thing talking to the database. Password isn't enough for a lot of applications. As long as vercel has lame security, I'll consider it a toy, nice for quick jokey blogs and stuff but not something to use for a serious project. Not about to risk getting sued just to use vercel s…

This is good feedback. Allowlist support is available for enterprise teams. We definitely hear you on VPC peering - it's on our roadmap.

More on the security of our platform here https://vercel.com/security, as well. Thank you!

Re: We are far from a better Heroku for production apps in a hyper cloud

#259

Post by leerob »

jopsen wrote: post #231
leerob wrote: post #165

Earlier quoted context omitted.

Hey! Lee from Vercel here. Vercel isn't just for static sites – it also handles server-rendered applications, serverless functions, and hybrid frameworks (like Next.js). The current focus has been on an incredible front-end experience, but we're working to become a first-party integrator with your favorite backend solutions. For example, you can deploy a full-stack Next.js application with a PostgreSQL database in le…

Isn't serverless still just lambda..? Slow to boot, and subject to denial-of-wallet attacks. I do love vercel for static sites, and also think my blog with a minimal backend for comments is a good fit. But less sure how it'll hold up to even a single heroku dyno, which can do a lot of concurrent request at very low cost.

Cold boots are going away with innovations like Firecracker[1]. One of our larger customers was able to handle 90 million concurrent requests during peak traffic.

[1]: https://www.serverless.com/blog/firecracker-what-means-serve...

Re: We are far from a better Heroku for production apps in a hyper cloud

#260

Post by jrochkind1 »

cortesoft wrote: post #191
jrochkind1 wrote: post #186

Earlier quoted context omitted.

I definitely wonder why someone else hasn't managed to compete with them on price for a similar and similarly high-quality service. But this post makes me think it's unlikely to be gitlab.

Because if you charged less you wouldn’t make enough to support a business?

It's possible.

But heroku charges a pretty big multiplier over the underlying AWS resources and does a pretty large volume (I don't know if the numbers are public, so I could be wrong, but my sense is they have a LOT of customers).

It seems likely to me that their prices are more about "what the market will bear" than about what they need to support the business. There's nothing wrong with that, nobody's forcing any customers to pay for heroku, the customers think the value is there.

It just makes me surprised that nobody's competing effectively on price. Which may mean I'm wrong and you are right. But there could also be other explanations.

If they HAD competitors offering the same thing, we might think that the competitors were forced to compete on price so all maybe were charging the least they could to support the business. At least that's the theory. But that they have no real competitors offering quite what they're offering... when what they're offering has been so successful... is surprising to me.

Maybe it's just hard to create something as high-quality as heroku Or to do it without funding, or to get funding for it?

Post reply on HN