Netbook Performance: Ubuntu vs. OpenSolaris

[kraftman]

@Kebbabert

If RCU is soo good for Linux, why does Linux scale bad on Big Iron, then? Admittedly, Linux scales well on large clusters, maybe because of RCU?

It scales great.

Anyway, RCU or not, Linux sucks on Big Iron. But scales well on clusters.

Linux scales great on Big Irons.

Question C) Can you prove that claim? Show me links proving that Linux is chosen over Solaris because of performance and stability issues. There are no such links. Why do you lie, then? Are you a liar?

Nope, you proof me it's not chosen, because of this.

Fine. If you want me to stop "spreading FUD" I will stop. I promise. But then I need to see some proofs so I can change my mind. Please show me links and proofs that you are correct, and I am wrong. If you can show such links, then I will change my mind. I promise.

I don't think so, because you're lying. I explained you many times some things and you still don't understand them. However, about Solaris stability (or rather no stability):

http://www.hardforum.com/showthread.php?p=1034455225

And you want to do this on Solaris?

Abso-fucking-loutely not. It's too unstable and unpredictable under loads. Not comstar; Solaris itself.

I've been working with Sun for nearly 20 years. Even attempting what you're talking about, is data suicide.

there's many more as you can read in this link.

Regarding your "valuable links", please do not show links on some guy having problems on installing ancient Solaris v8 - as a proof that Solaris is unstable on large systems. Please show me some real life testimonies. Do not show me a feature list or whatever.

I explained this few times. Still some problems? :>

http://svn.haxx.se/users/archive-2005-12/0507.shtml

Another example? There are many more which shows it sucks. Oh, be careful when using DTrace

http://en.securitylab.ru/notification/300509.php

Incredibly robust: http://forum.percona.com/index.php/m/4404/

And some design error: http://www.securityfocus.com/bid/9548

Que? What do you mean? Do you mean that professor and researcher Tannenbaum says that Linux scales bad? Or do you mean that Tannenbaum does not Linux? What the hell do you mean with this sentence? I CAN NOT READ YOUR MIND.

Tannenbaum didn't say Linux scales bad, so what are you talking about? I said don't fool yourself talking about Tannenbaum. I consider he's an asshole and maybe even an idiot. Very close minded one.

Ok, now you will face the problems I have with you. What do I mean when I write like this?
"Yes I know but I am correct on Tannenbaum, and your second argument is flawed"

You know I'm right in some part, but you still consider you're right when comes to some point. However, very bad example and in this case I suppose you didn't mean anything. This is a difference :> Btw. It's your problem. To understand me sometimes you've got to think, but you have problems with this.

[krisia2006]

Who'd like to run opensolaris on a netbook (except saying it's possible) ???
And for doing what ?

There are many more windows users who think the same thing about Linux. You are as narrow minded as they are...

[lordmozilla]

There are many more windows users who think the same thing about Linux. You are as narrow minded as they are...

Linux Community - Becoming as narrow minded as Apple fans.

[bnolsen]

Linux Community - Becoming as narrow minded as Apple fans.

Nah...windows is naturally an easy target. Technologically it's kludged together crap that sells due to inertia and marketing.

Apple and linux folks are very different animals, with the linux folks probably more likely wanting to kick the crap out of the apple folks if locked together in a room too long.

I really, really find it interesting how these solaris threads bring out the passion (and trolls) in people. I guess I was burned too much by solaris in the 90's with my experience with them being deployed both on the server and the desktop. At the time Sun was guilty of highway robbery marking up crap like external cd rom burners 5000% while providing inferior cpu performance compared with intel (starting even with the 486DX series). And my experience with Sun on the desktop showed how buggy their stuff was...

My point is that solaris provides no awesome wonderful magic bullet. I wouldn't trust ZFS near anything high performance/high volume. In the same way I for certain don't trust JFS or XFS, they also fall apart under certain conditions.

Thing about it is, solaris and linux aren't really so much different. I personally feel more comfortable with linux and feel its more accessible than solaris, but generally marginally so. They're in the same family and generally comparable. The OS's themselves I believe aren't as much a deciding factor as the community and culture around these products.

[Apopas]

Nah...windows is naturally an easy target. Technologically it's kludged together crap that sells due to inertia and marketing.

Apple and linux folks are very different animals, with the linux folks probably more likely wanting to kick the crap out of the apple folks if locked together in a room too long.

As far as I understand, Macosx users don't differ that much than windows ones. Marketing is the Apple's great weapon all these years just like Microsoft and not price, freedom and performance just like Linux. Furthermore, isn't rare to meet an Apple user that knows nothing about computers besides the few programs he uses. That's not the case for Linux users despite the community tries to approach such users today but it fails since only with raw great marketing you can persuade this type of folks to choose your product. And only MS and Apple can succeed to this. Nah, Linux and Apple users don't share anything in common except the fact that they dislike windows, but with Linux there is an ideological substratum along with personal preference that leads to that, while with Apple's that derives from strict personal preference mainly, rather than ideology.
That's why Linuxers will duel often for licenses while macers and windowsers won't care for the license at all but just if the program works.

[nanonyme]

As far as I understand, Macosx users don't differ that much than windows ones. Marketing is the Apple's great weapon all these years just like Microsoft and not price, freedom and performance just like Linux.

Psst, out of those three price is a badly defined concept (OEM prices) and freedom is uninteresting to most people (who are ready to give up even their civil rights to their governments to be more safe). So the only useful number is performance and to convince people your system is faster, you need to tell them that it is. Hence marketing.
If you want a system majority wants to use, it has to be fast, secure and easy to use all at once, and people need to hear and believe it. (well, just having people believe it is enough but always better that it actually is so it's less trivial for competitors to convince them otherwise) It's not nearly all about software engineering but also having people realize you and your product exist.

[Apopas]

Psst, out of those three price is a badly defined concept (OEM prices) and freedom is uninteresting to most people (who are ready to give up even their civil rights to their governments to be more safe). So the only useful number is performance and to convince people your system is faster, you need to tell them that it is. Hence marketing.

Yup, that's exactly my point. That's why Linux users (and the other FOSS of course) differ so much than the proprietary ones.
Linuxers and Macers has nothing in common. Windowsers and Macers share a lot.

[Svartalf]

Tannenbaum didn't say Linux scales bad, so what are you talking about? I said don't fool yourself talking about Tannenbaum. I consider he's an asshole and maybe even an idiot. Very close minded one.

I'd be cautious calling him either, kraftman. He's probably forgotten more about computers, computer science, and the like, than most of have learned over the years. Andy's one of the truly sharp ones out there. Now, if you were to say he was opinionated and sometimes wrong like all humans can be (even Linus is that... )- I'd concur heartily.

[kraftman]

I'd be cautious calling him either, kraftman. He's probably forgotten more about computers, computer science, and the like, than most of have learned over the years. Andy's one of the truly sharp ones out there. Now, if you were to say he was opinionated and sometimes wrong like all humans can be (even Linus is that... )- I'd concur heartily.

Yeah, I'm sometimes not fair :/ It makes me sick when someone tries to force some idea etc. but I shouldn't be calling him. Btw. I don't consider Linus being some kind of super human etc. Nobody is, in my opinion :>

[misiu_mp]

Here is a typical Linux FAIL scenario:

Customer has hard limit for transaction completion, say 5 seconds with 10000 users. Linux may complete 99.9% of transactions in 1.5 seconds, but 0.01% come back in 6 seconds, FAILING the test.

On same hardware, Solaris completes 100% of transactions within 5 seconds and PASSES the test. Even if the average response time is twice Linux, Solaris WINS and Linux LOSES.

Just to clarify things. The above describes a deadline problem. Guaranteeing deadlines is something that only real-time operating systems do.
As far as I know neither Solaris nor Linux is a real-time operating system, thus saying that any of those systems would complete some transaction within a certain deadline in 100% of cases is by definition untrue. Unless a system is designed to always meet deadlines (thus being a real-time system), it will not always meet deadlines.

If a customer REALLY has a HARD limit for transaction completion, such as in case of a laser cutter controller or a life-support system, the customer should choose a HARD real-time operating system. Not Solaris, not Linux (at least not the normal one, RTLinux is a commercially suported hard real-time microkernel that wraps around Linux and possibly could do the job).