Hacker News Reader

Comment by naikrovek | Hacker News Reader

Honestly, I don't know how the landscape looks today.

At the time, FreeBSD could run Linux binaries faster than Linux could on the same exact hardware, for one.

At Yahoo, we used FreeBSD on all of our servers and Linux on zero of our servers, and that probably changed over time. I don't know.

It was just very stable. You could reboot without worry that a kernel update would break anything or prevent boot-up. It was almost a realtime OS considering how light it was. I don't think it actually qualifies as an RTOS but we knew exactly what performance we could get out of a particular server configuration, and if the performance ever deviated, we knew there was a problem. It was very deterministic in that way.

Linux may be as good or better than FreeBSD now. I don't know. But 20 years ago, Linux getting all the attention saddened me a lot.

At the time, if you wanted an open operating system, without any cruft clogging things up, FreeBSD was it.

View on HN

When I started in 2004, Yahoo used FreeBSD everywhere except for acquisitions (including Inktomi which was a large part of the host count, running Linux); when I left in 2011, it was more of a mix; there wasn't yet a mandate to migrate to Linux, but I suspect it was coming. The mandate at that time was more of support both.

I went to another company that was using FreeBSD. They were three major versions ahead of what I had at Yahoo, and I didn't need to relearn very much, almost everything I knew still worked, but performance was improved.

>there wasn't yet a mandate to migrate to Linux, but I suspect it was coming.

I think it was Marissa Mayer that mandated the switch to Linux. Facebook also forced WhatsApp to switch to Linux after their acquisition.

This is all still true. FreeBSD is amazing as an operating system, everything is integrated nicely, everything just works, sure the userland isn't the same as GNU in some respects but you get used to it (and some of the GNU utilities are available if you want - like gtar).

Coming from the experience of trying to build from source on Linux a decade ago ports is just an amazingly streamlined experience, it all really just works.

I only run it on my fileserver but it's a really nice experience there. I don't have the experience to run it in production at work but it's really a nice setup.

AFAIK FreeBSD loses to Linux on benchmarks, so IDK why you are saying it runs Linux binaries faster than Linux itself.

Well it can still be true that FreeBSD runs Linux binaries better than its own native binaries, though I don’t think this is the case.

What was stated is that BSD runs Linux binaries better than Linux, not better than BSD binaries.

Anyway, I seriously doubt this is the case. Almost every single high performance computer runs on Linux. Most servers run Linux. I'd say except from Netflix, FAANGS run Linux almost exclusively for their servers.

I have a hard time believing FreeBSD is better than linux in any meaningful way, more so given the humungous resources The Linux Foundation has.

In the early 2000s, Linux binaries did run faster on FreeBSD than on Linux on the exact same hardware, when using a stock FreeBSD install and a stock (but light) Linux install.

I recall reading at the time that it was because FreeBSD's kernel simply had less overhead overall and consumed fewer resources in its work, making that bit of CPU available to running processes instead of consuming it itself.

After using FreeBSD, the Linux kernel appeared very bloated and slow, at the time. And it was, as distributed by distro makers. So the comparison was fair, stock FreeBSD vs. stock Linux. If you want to argue that you could do better than stock FreeBSD with a custom Linux kernel, you'd be right, and if you allowed a fair comparison with a custom FreeBSD kernel, I think Linux would have again struggled to remain in the lead.

All of that being said, this landscape may have changed entirely. The Linux kernel has come a long way. The FreeBSD kernel has come a long way. So where things are now is unknown to me.

I will say that superior performance is not always the reason to choose an OS for high performance computing tasks. It is a sane decision to use an OS that has lower performance if it provides other benefits that are important to you.

Above it's been claimed the performance difference still stands. I seriously don't believe so.

Then test and see for yourself, if you want to know for yourself. We aren't making these things up.

Benchmarks I've seen everywhere show worse performance for FreeBSD vs Linux.

Plus, if someone claims something, you bear the burden of the proof. If a person claims BSD runs Linux binaries faster than Linux nowadays, then that claim should be backed up by the truth.

I'm not gonna comment how was it 20 years ago because there seems to be quite an agreement on that, and I can't doubt BSDs where more mature than Linux at that point given Linux was younger.

It's not always faster or slower:

https://www.phoronix.com/scan.php?page=article&item=freebsd-...

And please stop trusting benchmarks, they say absolutely nothing until you make that exact benchmark that mirrors your real-world workload. Otherwise it's just nonsense (except to find regressions)

No just sometimes ;)

https://www.phoronix.com/scan.php?page=article&item=linux_ga...

A 10 year old benchmark full of questions and disagreements in the comments list. Si nce when is that a proof?

Since when is a synthetic benchmark proof at all?

Saying a OS is faster than X is like saying a ferrari is faster than a jeep..on perfect roads..yes, on bad terrain..no, not even usable.

I in fact did not say that it runs Linux binaries faster than Linux itself.

I confirm the fact that linux binaries were noticeably faster at least in 2000, when I was using it exclusively. Also I had a laptop with a broken memory module. Linux would panic unless I put it in the fridge (don’t ask me how this came about, but it helped), but freebsd was stable, apart from some crashing applications. And the kernel configuration was easy to read and understand.

Also if firewalling/packet filtering was a consideration, ipfw was (is?) leaps and bounds nicer to work with than ipchains/iptables. This (c. FreeBSD 4.x) is what introduced me to FreeBSD.

OpenBSD’s pf (which got ported to FreeBSD) and it’s pf.conf still sets the benchmark for the most beautifully readable firewall rules.

The FreeBSD pf is an (very) old version of what is available on OpenBSD.

No it's not, it's a fork.

The config file syntax matches that of the old pre-4.5 OpenBSD pf but the code under the hood has seen much evolution. FreeBSD's pf is SMP-capable, for one thing, unlike OpenBSD's that "will only use one processor": http://www.openbsd.org/faq/pf/perf.html

Reply on news.ycombinator.com