“The next major update of FreeBSD 7, due this December, is in the running to be one of the most impressive FreeBSD releases to date. The ULE scheduler has now reached maturity, leading to significant gains across the board (particularly in server workloads). This new scheduler brings notably impressive performance improvements to both MySQL and PostgreSQL. In the first section of this article, I’m going to take a look at what’s new. In the second section, I will discuss what the future holds for FreeBSD beyond the upcoming FreeBSD 7.0 release, including screen shots of the revamped FreeBSD installer ‘finstall’.”
Well… the review was a bit short but ok.
Here’s YAB (Yet Another Benchmark):
http://www.kernel.org/pub/linux/kernel/people/npiggin/sysbench/
Edited 2008-03-06 16:51 UTC
That’s why I called it a preview in the headline .
But it is not a preview of FreeBSD 7.0, as 7.0 is already released. The title of the original article says Review, not Preview.
The article is a nice preview of the features in the upcoming releases after 7.0, however.
Edit: But it seems indeed that the original article is a bit late and should have been published earlier, already before 7.0 was released, and the text is a bit confusing therefore.
Edited 2008-03-06 18:12 UTC
Well, your description is far better than most sites that turn up if you google “freebsd 7 review” including the site you linked to. “The next major update of FreeBSD 7, due this December”,”I’m certainly looking forward to the pending 7.0 release and beyond.”
In other news, “Hitler invades Poland” 3/7/2008.
>Linux kernel 2.6.25-rc4 was the Linux version tested
Maybe with the kernel release it will be even better, maybe it will be worse. Linux is always a developer snapshot since they abolished stable and developer kernels.
>The Linux kernel used is not a “stable release” whereas FreeBSD is (although I’m not aware of any significant performance improvements over the 2.6.24 kernel
–from the benchmark
But maybe there are heavy bugs in the .24 kernel?
>Maybe with the kernel release it will be even better, maybe it will be worse. Linux is always a developer snapshot since they abolished stable and developer kernels.
Get a grib, 2.6.x.x is stable releases.
C’mon this is no stable release, you should really nobody with a sane mind would use such kernel in production! Yeah apart from some desktop users (hello Windows) :->
Have a look at Debian, guess why they are using no bleeding edge kernel? Because _you_ and every other user of bleeding edge kernels are the beta tester!
But don’t be to tearful, there is a life after the flame
As you can see here, the ultimative truth about Linux and BSD *g*
http://www.fixmbr.de/beetlebums-bilder-zum-chemnitzer-linux-tag-200…
We can see, but we cannot read the funny(?) German comments in between.
Linux is great when everything works, has better hw support, and Debian is a great Desktop system most of the time, hail aptitude!, but when you have a problem:
BSD – How do I configure xxxx? man xxxx
Linux – How do I configure xxxx? man xxxx, huh? google xxxx, huh? google Fedora Core kernel 2.6.22 xxxx, “Hey xxxx sploded on me wtf omg i needs hlp”,”RTFM”,”recompile your kernel with option –xxxx”, huh?
Nobody has confirmed or verified this yet.
Kris’ benchmarks were up to date until FreeBSD 7 was released He cannot test for future releases! Oh one more thing If you u like to compare new stuff (like 2.6.25-rc4) then you must try FreeBSD 8 So, let’s not rush like Slashdotters by submitting “an update” as if Linux is going to die because of this.
Of course, such challenges between open source devs is always welcome, it will delivier better products. But if Piggin is just trying to kill the news about FreeBSD then he’s just wrong. Jeff and Kris has been in contact with Linux devs and discussed these issues for the past year to improve and point out weaknesses. So, when someone just appears in one day to prove that Linux is l33t and better is just …..
Edited 2008-03-07 05:45 UTC
Hmmm. That shows that Linux is really 33% faster than FreeBSD 7.0. But the FreeBSD press release said that FreeBSD 7.0 is 15% faster than the fastest Linux. I’m confused! 😉
Nobody has confirmed or verified this yet.
Kris’ benchmarks were up to date until FreeBSD 7 was released He cannot test for future releases! Oh one more thing If you u like to compare new stuff (like 2.6.25-rc4) then you must try FreeBSD 8 So, let’s not rush like Slashdotters by submitting “an update” as if Linux is going to die because of this.
Of course, such challenges between open source devs is always welcome, it will delivier better products. But if Piggin is just trying to kill the news about FreeBSD then he’s just wrong. Jeff and Kris has been in contact with Linux devs and discussed these issues for the past year to improve and point out weaknesses. So, when someone just appears in one day to prove that Linux is l33t and better is just …..
some linux dude modded me down I am just reposting so others do not ask this question again.
Edited 2008-03-07 10:43 UTC
But according to this, 2.6.22, released 8 months ago, also trashes FreeBSD 7.
What is the planned release date of FreeBSD 8 again? I can’t quite remember what the plan is on that. 2.6.25 is planned for sometime between April and June.
Edited 2008-03-07 11:24 UTC
According to Who? what trashes what? what you talking about?
http://people.freebsd.org/%7Ekris/scaling/os-mysql.png
Btw the image timestamp is 2007-Nov-19
You do not have to wait for it. Just csup HEAD and volla you will get 8 with increased performance
All I am trying to say that the slides clearly indicated what version and release they were comparing to. And obviously these were the current versions of linux at that time and linux devs are aware of it and they improved so many things after that. Now all of a sudden Piggin wakes up and tries 2.6.25-rc4 to tell us that the new merged patches improved the performance of linux after all? What is this highschool? or what I got bigger muscles that yours typa thing? Why can’t you just admit that FreeBSD devs worked hard for the past 5 years to get this done and deserve it? Not oh see see see I got better results now then “LATEST NEWS UPDATE ON SLASHDOT” by an anonymous submitter! FreeBSD performance does not even come close to 2.6.25-rc4.
If you want to compare stuff that is not released then perhaps you have to be fair and use freebsd 8 (the latest in the tree)
Edited 2008-03-07 12:06 UTC
Benchmarks are like a$$holes; everyone has one and your life will be better if you don’t look at other’s.
Ummm FreeBSD 7.0 is not coming out ‘this december’ it’s actually already out now.
Check out FreeBSD.org to download.
It seemed to me like this was no a very good “advertisement” for FreeBSD 7.0. Here’s how I read the story:
1. New things coming:
——ZFS – cool, except features A,B and C don’t works, and it’s still beta quality
——DTRACE – cool, except it really unstable, probably not ready for production
——A new GUI install
The rest seems to be mostly code cleanup and better threading.
Mind you, that’s just how it read to me. Personally, I think FreeBSD rocks, and I believe 7.0 will really shine – especially due to the improved SMP support in this day of ubiquitous multi-core processing.
Who needs advertisements on this site’s news section anyway? Isn’t it good to be open and honest also about the things that still should be improved? At least I’ve had enough of empty advertisement journalism and fanboyism. May the quality of the technology speak for itself. For some reason; I’ve also come to the conclusion that the BSD world seems to be more full of fanboyism than the Linux world (although also the latter one sure has a fair share of it).
As to the article, I didn’t find anything negative in it. On the contrary, the writer seems to be very enthusiastic about the new and upcoming features in FreeBSD.
It’s a nice article, but definitely a bit out of date:
It’s not “upcoming”. we’re beyond that already.
Thom.. do you even read the headlines of the news you post..? If not, read the ones you posted a week ago
When I wrote and published this FreeBSD was supposed to be released end of December — of course the deadline was missed / pushed back as it almost always is.
The preview/review was of the RC releases back in the day (2 months ago I believe).
I think this was published in Jan or Feb when the text was slightly less confusing.
An avid OS News reader.
-Yousef Ourabi
Release date aside i agree an interesting review. On the FreeBSD.org website there is a link for one of the few ‘news’ stories on the FreeBSD 7 release.
Not much in the way of deep feature review but kinda interesting to get some FreeBSD developers comments about who they compete against and other businessy kind of things.
here’s a link : http://www.internetnews.com/dev-news/article.php/3731386