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

Ferron – A fast, memory-safe web server written in Rust

github.com

Post reply on HN
Print view View original

51–60 of 102 posts

Re: Ferron – A fast, memory-safe web server written in Rust

#51

Post by MaxBarraclough »

austin-cheney wrote: post #36

Every web server claims to be fast, so I wonder how they define that. As someone who has written their own supposedly fast web server I only want configuration simplicity. Most web servers are unnecessarily far too complicated. In a web server here is what I am looking for: * Fast. That is just a matter of streams and pipes. More on this later. That said the language the web server is written in largely irrelevant to…

> Fast. That is just a matter of streams and pipes. More on this later. That said the language the web server is written in largely irrelevant to its real world performance so long as it can execute low level streams and pipes.

I'm no expert, but that doesn't sound right to me. Efficiently serving vast quantities of static data isn't trivial, Netflix famously use kernel-level optimisations. [0] If you're serious about handling a great many concurrent web-API requests, you'll need to watch your step with concerns like asynchrony. Some languages make that much easier than others. Plenty of work has gone into nginx's efficiency, for example, which is highly asynchronous but is written in C, a language that lacks features to aid with asynchronous programming.

If you aren't doing that kind of serious performance work, your solution presumably isn't performance-competitive with the ones that do. As you say, anyone can call their solution fast.

[0] [PDF] https://freebsdfoundation.org/wp-content/uploads/2020/10/net...

Re: Ferron – A fast, memory-safe web server written in Rust

#53

Post by dorianniemiec »

nottorp wrote: post #52

Isn't Go better for writing servers, and as fast and memory safe as the second coming of $DEITY?

Go has larger ecosystem of libraries for building web servers. You have FrankenPHP for running PHP, Lego for automatic TLS, etc. For Rust there is `tokio-rustls-acme` crate (used by Ferron) for automatic TLS. While for PHP there is a `php` crate that depends on unsupported PHP version. Ferron uses FastCGI for communicating with PHP-FPM daemon instead. However, Go uses a garbage collector, unlike Rust, which has a borrow checker to ensure memory safety.

Re: Ferron – A fast, memory-safe web server written in Rust

#54

Post by rc00 »

dorianniemiec wrote: post #53
nottorp wrote: post #52

Isn't Go better for writing servers, and as fast and memory safe as the second coming of $DEITY?

Go has larger ecosystem of libraries for building web servers. You have FrankenPHP for running PHP, Lego for automatic TLS, etc. For Rust there is `tokio-rustls-acme` crate (used by Ferron) for automatic TLS. While for PHP there is a `php` crate that depends on unsupported PHP version. Ferron uses FastCGI for communicating with PHP-FPM daemon instead. However, Go uses a garbage collector, unlike Rust, which has a bor…

Rust has garbage collection.

Re: Ferron – A fast, memory-safe web server written in Rust

#55

Post by 1dom »

austin-cheney wrote: post #36

Every web server claims to be fast, so I wonder how they define that. As someone who has written their own supposedly fast web server I only want configuration simplicity. Most web servers are unnecessarily far too complicated. In a web server here is what I am looking for: * Fast. That is just a matter of streams and pipes. More on this later. That said the language the web server is written in largely irrelevant to…

Can you share a link to your web server please?

I'm finding it hard to make sense of your comment: I can't reconcile some of the stuff you're saying. My gut feeling is you're either ridiculously smart, so smart that defining and implementing a security rules engine for a web server is something genuinely trivial for you, and the world has a lot to learn from you. Or, you're really, really not aware of how much you don't know, so much so that you're going to end up doing something stupid/dangerous without realising.

Either way, an example of a supposedly fast web server written by you should clear it up pretty quickly.

Sorry, because this feels a little rude, and I don't mean it to be, but you're contradicting quite a lot of widely held common sense and best practise in a very blasé way, and I think that makes the burden of proof a little higher than normal.

Re: Ferron – A fast, memory-safe web server written in Rust

#56

Post by dorianniemiec »

rc00 wrote: post #54
dorianniemiec wrote: post #53

Earlier quoted context omitted.

Go has larger ecosystem of libraries for building web servers. You have FrankenPHP for running PHP, Lego for automatic TLS, etc. For Rust there is `tokio-rustls-acme` crate (used by Ferron) for automatic TLS. While for PHP there is a `php` crate that depends on unsupported PHP version. Ferron uses FastCGI for communicating with PHP-FPM daemon instead. However, Go uses a garbage collector, unlike Rust, which has a bor…

Rust has garbage collection.

How? I rather think that it uses a borrow checker with ownership and borrowing rules.

Re: Ferron – A fast, memory-safe web server written in Rust

#57

Post by sim7c00 »

1dom wrote: post #55
austin-cheney wrote: post #36

Every web server claims to be fast, so I wonder how they define that. As someone who has written their own supposedly fast web server I only want configuration simplicity. Most web servers are unnecessarily far too complicated. In a web server here is what I am looking for: * Fast. That is just a matter of streams and pipes. More on this later. That said the language the web server is written in largely irrelevant to…

Can you share a link to your web server please? I'm finding it hard to make sense of your comment: I can't reconcile some of the stuff you're saying. My gut feeling is you're either ridiculously smart, so smart that defining and implementing a security rules engine for a web server is something genuinely trivial for you, and the world has a lot to learn from you. Or, you're really, really not aware of how much you do…

a rules engine for a web server isnt difficult if it doesnt have to carry responsibility for the web app security itself. then its as the posted outlined really...

that being said, its common for new servers to have old vulns, not many coders will go over cve reports of apache and nginx and test their own code against old vulns in those.

i do find a lot of claims about performance or security are oftend unsupported as with this server. it just says it on the readme but provides no additional contex or proof or anything to back up those claims.

my thought is that original commenter gott riggered by that, perhaps rightfully, and points out thia fact more than anything. if you want to claim high performance or security, back it up with proof.

the simple fact its in rust doesnt make it more secure. and using async in rust doesnt imply good performance. it could in both cases. wheres the proof.?

Re: Ferron – A fast, memory-safe web server written in Rust

#58

Post by dorianniemiec »

sim7c00 wrote: post #57
1dom wrote: post #55

Earlier quoted context omitted.

Can you share a link to your web server please? I'm finding it hard to make sense of your comment: I can't reconcile some of the stuff you're saying. My gut feeling is you're either ridiculously smart, so smart that defining and implementing a security rules engine for a web server is something genuinely trivial for you, and the world has a lot to learn from you. Or, you're really, really not aware of how much you do…

a rules engine for a web server isnt difficult if it doesnt have to carry responsibility for the web app security itself. then its as the posted outlined really... that being said, its common for new servers to have old vulns, not many coders will go over cve reports of apache and nginx and test their own code against old vulns in those. i do find a lot of claims about performance or security are oftend unsupported a…

Regarding the server performance, there are benchmark results on the Ferron's website.

Re: Ferron – A fast, memory-safe web server written in Rust

#60

Post by graemep »

dijit wrote: post #15
alexpadula wrote: post #13

Earlier quoted context omitted.

You’re comparing a new project to nginx. Obviously nginx will be faster maybe not across the board but generally it probably is. As a project matures it will optimize surely! nginx has 21 years of development under its belt.

By that reasoning Apache should be faster than nginx, but alas. https://pressable.com/blog/head-to-head-performance-comparis... I think drawing any conclusions in the absence of benchmarks is unwise.

It is often a mistake to draw conclusions even with benchmarks. Are the benchmarks measuring what is relevant to your use case? Are the benchmarks unbiased.

Your link does not seem to contain any benchmarks anyway.

The Ferron benchmarks on their home page say Apache Pre fork MPM outperforms Apache Even MPM which seems odd to me.

Post reply on HN