Woah, It Looks Like Oracle Will Stand Behind OpenSolaris

[kraftman]

A source on what? Could you be more specific? I don't know how many times I have told you "what is clear to you, is not clear to me". But I just don't reach you on this. I have explained that many times. But you just don't understand. It is tiresome.

For this: Larry has invested billions in Solaris, far more money than Linux. You tried to explain many things many times, but you've got problems in proving your claims.

I don't care how many amateur devs Linux have. Nothing new is invented in Linux. Linux devs just copy: ZFS, DTrace, etc. And the copies are bad. Ok if the copies where better, but no, all copies suck bad. Not to mention how bloated and buggy Linux is - as Linux kernel devs say. Including Linus T. Solaris dont have that bad reputation that Linux has.

I don't care about amatours. ZFS and Dtrace are the only things unique to Solaris? Oh, they aren't unique. ZFS is a file system and dtrace is some tool. Those are COPIES. In this case, what Linux copied is better. Not to mention how bloated and buggy Solaris is, as some articles says (even quite old). Linus said Linux code quality is getting better and better, so... There's also some fine site:

http://scan.coverity.com/rung1.html

It shows Linux code quality is awesome. Much better then NetBSD. Sadly, there's no Solaris and FreeBSD out there. Maybe Solaris guys are afraid or something? Slowlaris has a very bad reputation - slow, bloated, support very limited number of hardware etc. Even some sun paid morons claim it's slower on x86, 64bit architectures.

Oh yes? Do you have a source on this?

OpenSolaris guys said something like they will give up etc. For me, it's logical thing Solaris will die soon.

I have done the informal MENSA test on their web site. I mailed to MENSA and asked about my result, and they told me I am on the verge of getting accepted to MENSA. (I had 13 out of 18, and some MENSA members had 12 out of 18 on the informal test). Hence, my IQ is / or close to 131. This is far above IQ of 75. MENSA said they have many mathematicians and physicists as members.

And what is MENSA? Isn't this a group of people who are good in resolving MENSA tests? You should give back about 51 points, because you believe in MENSA, so you've got 75.

I was actually referring to other people, those who don't understand what one write. You know, when I write something many times, and he just doesn't understand. Then he seem to be quite dumb, don't you think? But I doubt there exist so dumb people here who doesn't understand, no matter how many times I explain. No one can be so dumb to have an IQ of below 75.

You're the one who don't understand, no matter how many times someone will explain, so I was right saying you should give 51 points back.

[kraftman]

I don't care how many amateur devs Linux have.

Linux has FAR MORE specialists then Solaris can ever dream about.

[kraftman]

Oh, zfs copied some feature from XFS:

http://scalability.org/?p=389

ZFS has captured significant interest (in part due to Sun’s massive over-hyping of it). It combines low level disk block managers with volume management with file systems. It claims it is the first to do this, though if you look at the documents (man pages) for xfs, you realize that xfs had hooks for this (without the layering violations) long long ago on IRIX. That said, I don’t remember using this “feature” under Irix at all.

It's slowlaris that copies from others and from Linux.

And some truth about it:

http://scalability.org/?p=388

It's an obsolete.

[smitty3268]

How is it FUD? So what new hot cool tech has Linux got, that everyone wants? You tell me. I have never hear about anything new hot tech Linux devs create. They just copy. Actually, the whole Linux is a copy, a bad reimplementation of Unix. But on the other hand, many wants to (and already have, including Linux) copy Solaris tech. So is it a fact, or is it FUD?

FUD. How about TTM/KMS/all the graphics work going on - that's something no other Unix system is advancing, and it's much more of a killer feature for me than another filesystem. Or KVM - as far as I know, it was the first virtualization code to ever directly take advantage of being within a kernel, that's a lot more innovative than a filesystem or a debugging/tracing application, of which there are many even if they aren't quite as polished as the ones Solaris came out with.

If you say that high end Enterprise Unix such as Solaris has a bad reputation as buggy and bloated, then I want to see some links. Show us some links. Go ahead. You know, Solaris code is out there. And if someone found it to be ugly, then we sure would have heard it.

I'm just saying that on normal systems (not servers that ship with Solaris) Solaris tends to have many more driver compatibility issues than Linux does. And that's where the vast majority of kernel bugs are, it's pretty rare to have a crash come from the memory management code or scheduler, or somewhere else. If you're spending $20,000 on a machine that comes with Solaris, then I'm sure the drivers work perfectly on that machine. As far as bloated, freely admit i just made that up. Prove to me that Linux is bloated. I'm pretty sure I can compile a linux kernel to be smaller than the default one that comes with Solaris. Am i wrong? Your quote about Linux being bloated is just ridiculously misleading, for reasons that have gone over again and again. It's pure FUD.

Yes, if someone says things about me that is not true, I must correct them. Dont you do that? If I claim you "said Linux sucks", then you surely want to correct me. So how is my reaction different from yours?

Sigh...

Because normal people don't respond to "You're stupid" with "MENSA says I'm not". They either go "Nuh uh", or "No, you're stupid", or just ignore it as standard internet trolling. It's like the most cliched response i could possibly imagine.

[shmerl]

Don't forget, Oracle invests for example in BTRFS and other system stuff in Linux too.

[Ex-Cyber]

Or KVM - as far as I know, it was the first virtualization code to ever directly take advantage of being within a kernel

I'm not sure what you mean by "directly take advantage of being within a kernel", but KVM's kernel bits can be built as modules, and there were virtualization modules years before KVM. Aside from that, I don't know the details but I'm pretty sure that IBM has Linux beat by several decades on the virtualization front, just not on PCs.

[kebabbert]

For this: Larry has invested billions in Solaris, far more money than Linux.

http://www.computerworld.com/s/article/9131829/Oracle_s_Sun_buy_Ellison_praises_Solaris_thumbs_nose_at_IBM

"In explaining his decision to make the acquisition today, Oracle CEO Larry Ellison boiled down Sun into a company of two key assets: Java and the Solaris operating system."

Larry wanted Sun for the two key assets: Java and Solaris. He bought Sun for 7.4 billion USD. He has invested several billions to just get Solaris. And he also officially said he will invest even more resources than Sun ever had on Solaris.

Oracle Database is run on Solaris on more than any other OS. The reference platform OS of choice, for Oracle Database is Solaris - this was official and outspoken long before Oracle bought Sun. Oracle Database is developed on Solaris, and then ported to other OSes (maybe because of DTrace. Firefox developers have switched from Ubuntu to Solaris just because of DTrace:

http://blog.mozilla.com/rob-sayre/2007/09/06/land-of-dtrace/

Long before Oracle bought Solaris, the database and Solaris was the best combination. There are very strong reasons for Larry to buy Solaris and to continue to invest in Solaris. If you want to run the Enterprise High End Oracle Database, then the best choice is Solaris.

And also, if you want to run SAP for instance, then the best choice is Solaris. As I have shown earlier, Solaris beats Linux on SAP benchmarks, even using slower hardware. And probably there are many other benchmarks showing that Solaris and any Enterprise software is best combination.

You tried to explain many things many times, but you've got problems in proving your claims.

Know you are just FUDing about me. You CONFESSED earlier that you FUD. You admitted it. You wrote that you did FUD.

If I have problem proving my claims, then quote me. Show me old posts where I have problems. If you can not prove this, then you are FUDing again. But this time you FUD about me.

I don't care about amatours. ZFS and Dtrace are the only things unique to Solaris? Oh, they aren't unique. ZFS is a file system and dtrace is some tool. Those are COPIES.

You dont seem to understand. As usual. DTrace is something new and unique, there has never been possible to do the things DTrace allows, earlier. It is just not a copy. On the other hand, Linux copies DTrace (sucky Systemtap). IBM copies DTrace (ProbeVue). Mac OS X has ported DTrace. FreeBSD is just now, porting DTrace. Many OS wants DTrace. If it is so bad as you think, why do they copy and port it?

Not to mention ZFS. Everyone wants it too by porting or copying. Many say it is the best filesystem out there. ZFS has new unique features that no filesystem has before. For instance, ZFS protect your data against silent data corruption. Researchers have stressed ext3, XFS, JFS, ReiserFS, etc - and these filesystems can not protect your data. Your data is not safe. But ZFS is unique - it completely protects your data. I have shown you research papers on this before, and I can post them again if you wish.

Why are the Linux community so upset about CDDL incompatibility with GNU? Just because Linux can not get ZFS nor DTrace. If Linux community did not care about DTrace nor ZFS, they would not get angry. But they where very angry on this.

Even if you think Solaris tech suck, there are lots of Linux developers that wants Solaris tech. They do not agree with you, they think it is very good. Some say it is best in the market.

In this case, what Linux copied is better.

What is better? BTRFS which a RedHat developer called "broken by design"? The copy Systemtap which another Linux developer called "crap"? You want to see links on this?

Not to mention how bloated and buggy Solaris is, as some articles says (even quite old).

You have said this many times. I have always asked you to back up your claims and show those links you are talking about. But you have never showed any links that say Solaris is bloated and buggy. Never. I have asked you many many times. Unless you can show me any link, then you are just FUDing again.

Maybe you should stop FUD? It is much easier to say things that has some substance, things that you can prove with white papers and benchmarks.

Linus said Linux code quality is getting better and better, so...

Well, Linus said recently that Linux was bloated. Other Kernel devs have said that Linux code is buggy. It seems that Linux code is quite bad, then it can only be better. If you are at the bottom, it can only be better. Linux code is at the bottom. The problem is, it stays at the bottom. Just look at what all the Linux kernel devs says! They all say it is buggy / bloated, etc. You want to see links on this, so you can see I do not make this up?

There's also some fine site:

http://scan.coverity.com/rung1.html

It shows Linux code quality is awesome.

On your link Andrew Morton says
?COVERITY'S STATIC SOURCE CODE ANALYSIS HAS PROVEN TO BE AN EFFECTIVE STEP TOWARDS FURTHERING THE QUALITY AND SECURITY OF LINUX.?
And he has complained a lot about the bad quality of the Linux code. So I doubt your page showing that Linux code is super. It is not super, not even Andrew thinks it. That page only shows that Linux code is IMPROVING. Not it is super.

Much better then NetBSD.

Good for Linux.

Sadly, there's no Solaris and FreeBSD out there. Maybe Solaris guys are afraid or something?

Or maybe there are some other reason? You know, to have few bugs in the code, does not mean it is good quality.

Slowlaris has a very bad reputation - slow, bloated, support very limited number of hardware etc.

Prove it. Shows us links. Or you are FUDing again. I have asked you many times to show links on your claims, especially how "bloated and buggy Solaris is", but never have you showed one single link on this.

Even some sun paid morons claim it's slower on x86, 64bit architectures.

Actually, I would not be surprised on this one. It may be true.

x86 instruction set is very bloated and buggy. You want to see links on this? So, if you compared Solaris on x86 and on SPARC on equally fast CPUs, I suspect SPARC wins. SPARC is better, clock for clock than x86.

But, x86 is now much higher clocked than SPARC, and x86 has many cores. So Solaris runs faster on x86 now. There is no doubt on this. Everyone agrees on this, even Sun people. The fastest Solaris machines now, is x86. But x86 instruction set is really bad.

OpenSolaris guys said something like they will give up etc. For me, it's logical thing Solaris will die soon.

You must make a difference between OpenSolaris and Solaris. Larry has bought Sun for many billions, to get Solaris. It is not logical that Larry kills off Solaris. If you think that, you are not logical.

On the other hand, IBM has officially said they will slowly kill off AIX in favour of Linux. IBM is migrating from AIX to Linux. AIX will die. Solaris will not.

OpenSolaris is like Fedora, it is the new version of next gen Solaris 11. OpenSolaris code will be forked off, polished and debugged and released as Solaris 11. OpenSolaris will not die, because otherwise there will be no Solaris 11.

And what is MENSA? Isn't this a group of people who are good in resolving MENSA tests? You should give back about 51 points, because you believe in MENSA, so you've got 75.

Hahaha! You are just saying this, because you are not logical (which your reasonings show) and you most probably get very bad scores at IQ tests.

You're the one who don't understand, no matter how many times someone will explain, so I was right saying you should give 51 points back.

I have asked you many times to back up your claims, and show links. And I have asked you many times to be clearer. Sometimes you just write weird things, difficult to understand. Such as when you post things like "I have shown you this many times earlier" - what is "this"? What are you refering to? I have asked you to be clearer and refer back to the things you talk about, but to no avail. It seems you dont understand.

[kebabbert]

Linux has FAR MORE specialists then Solaris can ever dream about.

Maybe, but they are amateurs. Solaris few developers always create new cool tech. Such as ZFS and DTrace, etc. What does Linux devs do? Nothing interesting (except for gamers which shows Linux is more tailored to desktops than Enterprise Server Halls). And they write buggy code.

[kebabbert]

Oh, zfs copied some feature from XFS:

http://scalability.org/?p=389

It's slowlaris that copies from others and from Linux.

You know, it is a big difference to just have hooks and to actually do it. To "have hooks" is just a vision. To "have hooks" means the code is empty, but ready for new functionality. But the actual coding can make it or break it. If the code is bad and buggy and bloated, then everything breaks down. Then ZFS would suck, but it does not.




I heard BTRFS also has a layering violation? Is this correct? If this is correct, then that makes BTRFS also bad?




BTRFS seems to be ?broken by design?? A RedHat developer writes http://lkml.org/lkml/2010/6/18/144

?In the meanwhile I confirm that Btrfs design is completely broken: records stored in the B-tree differ greatly from each other (it is
unacceptable!), and the balancing algorithms have been modified in insane manner. All these factors has led to loss of *all* boundaries holding internal fragmentation and to exhaustive waste of disk space
(and memory!) in spite of the property ?scaling in their ability to address large storage?.
?
It seems that nobody have reviewed Btrfs before its inclusion to the mainline. I have only found a pair of recommendations with a common idea that Btrfs maintainer is ?not a crazy man?. Plus a number of
papers which admire with the ?Btrfs phenomena?. Sigh.

The first obvious point here is that we *can not* put such file system to production.?





The data security of ZFS is actually the only point in using ZFS. CERN did a major study on their precious data from the large particle machines, and all data gets bad over time. The hardware did not even notice this! CERN is now migrating from Linux hw raid to Sun ZFS machines.

http://storagemojo.com/2007/09/19/cerns-data-corruption-research/

Read the very bottom here

http://blogs.sun.com/simons/entry/hpc_consortium_big_science_means

Also, computer science researchers have injected errors into XFS, JFS, ReiserFS, ext3, NTFS, etc and all errors were not corrected. They where not even detected. There is a link to the research paper here:

http://www.zdnet.com/blog/storage/how-microsoft-puts-your-data-at-risk/169

In another comp sci study, ZFS detected all errors, and would have corrected them all if it used ZFS raid. In the study only a single disc where used so there where no redundancy to correct the detected errors.

http://www.cs.wisc.edu/wind/Publications/zfs-corruption-fast10.pdf

This data safety (which no other filesystem offers) is the ONLY reason to use ZFS. Never mind the fancy functionality. You dont want to corrupt your valuable data. And to get the checksums correct, is very very very difficult to do correctly. Look at all XFS, JFS, etc ? they do not succeed in correcting nor detecting errors ? and they compute checksums all the time.

And some truth about it:

http://scalability.org/?p=388

It's an obsolete.

That guy is totally off. He says that Solaris is bad for not having many drivers. That is weird. Solaris is for large enterprise halls, not for desktop use with fancy graphics and such. Solaris main concern is stability, a Server OS. Not desktop. There are other priorities in Server OSes. He dont understand that.

He is talking from Desktop perspective, and then yes, Solaris is not as good desktop as Linux is. But, if he talked about Server perspective, then Linux sucks badly. No scalability, bad uptime, bad patching (patch something and you break something), etc.

[kebabbert]

FUD. How about TTM/KMS/all the graphics work going on - that's something no other Unix system is advancing,

The other Unixes are not interested in desktop things. They are Enterprise Server OSes. Designed for mission critical things. Not high FPS in Crysis. Look, Linux is a desktop OS trying to get into the server halls. Solaris is a server os, trying to get to desktops.

and it's much more of a killer feature for me than another filesystem.

I bet you think the best OS is the one who gets highest FPS in Crysis? Desktop user.

Or KVM - as far as I know, it was the first virtualization code to ever directly take advantage of being within a kernel,

Maybe if you talk about Linux kernel. But other Unix have had similar functionality for a long time.

that's a lot more innovative than a filesystem or a debugging/tracing application, of which there are many even if they aren't quite as polished as the ones Solaris came out with.

DTrace, ZFS etc is not about polish. They are unique, and they offer new things no one has ever been seen before. DTrace is totally unique. Nothing comes close. DTrace is revolutionizing unique. It is just not "more polished" - which shows you are not a developer. ZFS offers data safety, which no other common filesystem has ever offered before. Read my long post above.

Look, if you think Solaris tech is just slightly more polished than the Linux copies - then I understand your view point. But that is not the case. Solaris offers completely new unique stuff no one has ever offered before. They revolutionize everything. THAT is the reason Linux devs so badly want Solaris tech. Not because it is "more polished". You should learn more about this before you down talk it. It only shows your ignorance.

I'm just saying that on normal systems (not servers that ship with Solaris) Solaris tends to have many more driver compatibility issues than Linux does.

This is true and I agree with you on this. But Solaris is a server os, not a desktop OS. Now it is trying to get to desktops.

And that's where the vast majority of kernel bugs are, it's pretty rare to have a crash come from the memory management code or scheduler, or somewhere else.

So all Linux bugs that Linux kernel devs talk about, are in the drivers? Jesus. You have not heard about, for instance, when Linux overcommits RAM by default? That is horrific and no server should use such a badly designed OS.

If you're spending $20,000 on a machine that comes with Solaris, then I'm sure the drivers work perfectly on that machine. As far as bloated, freely admit i just made that up.

It is good that you admit you made it up, that "Solaris is bloated". But it is not.

Prove to me that Linux is bloated. I'm pretty sure I can compile a linux kernel to be smaller than the default one that comes with Solaris. Am i wrong? Your quote about Linux being bloated is just ridiculously misleading, for reasons that have gone over again and again. It's pure FUD.

I am refering to when Linus T said that "Linux is bloated". That is proof enough.

Because normal people don't respond to "You're stupid" with "MENSA says I'm not". They either go "Nuh uh", or "No, you're stupid", or just ignore it as standard internet trolling. It's like the most cliched response i could possibly imagine.

I am not trolling. I have never made anything up. Everything I say, I can show links and white papers and benchmarks. You, on the other hand, have confessed you made things up. I do not. Kraftman has confessed he FUDs.

Look, I am not saying Linux is bad per se, it is actually a good OS. But in comparison with Mainframes, OpenVMS, high end Enterprise Unix, it is bad. But it is catching up. In the future, maybe Linux will catch up. Then I will stop quoting Linux kernel devs that say that Linux code quality is bad, is bloated, etc. I just quote them. I do not make things up. I can show links if you wish.