Torvalds: Solaris is a Joke
Q:
What do you think of what Sun is doing with Solaris 10--technology
improvements, open source, and the move to x86 chips?
A: I'm taking a very wait-and-see attitude to Sun. They like talking too much. I'm waiting for the action.
...
Q: It seems to me that they have taken some action besides just grandstanding. They have resurrected the x86 version and added several interesting features--containers, DTrace, and ZFS, for example--that are available today in beta versions of Solaris 10. They're actively rounding up support from developers and software companies. And they announced that the production version of Solaris 10 on x86 will be available for free. What do you think about the x86 move and the new Solaris features?
A: Solaris/x86 is a joke, last I heard. (It has) very little support for any kind of strange hardware. If you thought Linux had issues with driver availability for some things, let's see you try Solaris/x86.
...
Q: When Sun releases Solaris as open-source software, will you take a peek?
A: Probably not. Not because of any animosity, but simply because I don't have the time or the interest. Linux has never been about "others," it's been about getting better than itself, so I don't really have any motivation to play around with Solaris. I'm sure that if it does something particularly well, people will be more than happy to tell me all about it.
...
Q: Surely if you like the idea of standing on the shoulders of giants, there might be some handy ideas in Solaris. Why ignore it?
A: Because I personally don't think they have anything left worth taking after I've applied the general Unix principles. I really do think Linux is the better system by now, in all the ways that matter.
But more importantly, if I'm wrong, that's OK. People who know Solaris better than I do will tell me and other people about the great things they offer. To try to figure it out on my own would be a waste of time.
A: I'm taking a very wait-and-see attitude to Sun. They like talking too much. I'm waiting for the action.
...
Q: It seems to me that they have taken some action besides just grandstanding. They have resurrected the x86 version and added several interesting features--containers, DTrace, and ZFS, for example--that are available today in beta versions of Solaris 10. They're actively rounding up support from developers and software companies. And they announced that the production version of Solaris 10 on x86 will be available for free. What do you think about the x86 move and the new Solaris features?
A: Solaris/x86 is a joke, last I heard. (It has) very little support for any kind of strange hardware. If you thought Linux had issues with driver availability for some things, let's see you try Solaris/x86.
...
Q: When Sun releases Solaris as open-source software, will you take a peek?
A: Probably not. Not because of any animosity, but simply because I don't have the time or the interest. Linux has never been about "others," it's been about getting better than itself, so I don't really have any motivation to play around with Solaris. I'm sure that if it does something particularly well, people will be more than happy to tell me all about it.
...
Q: Surely if you like the idea of standing on the shoulders of giants, there might be some handy ideas in Solaris. Why ignore it?
A: Because I personally don't think they have anything left worth taking after I've applied the general Unix principles. I really do think Linux is the better system by now, in all the ways that matter.
But more importantly, if I'm wrong, that's OK. People who know Solaris better than I do will tell me and other people about the great things they offer. To try to figure it out on my own would be a waste of time.
That said, Linus needs to remember that without the open source community, Linux would be a joke. Sun's move to open source S10 is designed to help address problems like the lack of driver support. I've been at Sun long enough to know that whenever the "competition" criticizes us, it's a good thing. It means we're doing something right.
Posted by Kevin on December 22, 2004 at 02:27 AM JST #
Posted by SS on December 22, 2004 at 02:35 AM JST #
Most of the things I know about solaris 10 was from public places (sun.com, bigadmin, blogs.sun, etc) but, I still go to customers talking about Solaris 10 and, most of them, know nothing about Sol-10 because they don't have the time or the need to look for every press release Sun issues. I think you (we?) should see Linus comment has a warning that our job isn't done, we MUST continue to push our efforts in passing our message across.
Posted by Jaime Cardoso on December 22, 2004 at 02:45 AM JST #
First of all Solaris is about stability and scalability. Solaris currently scales to 100+ cpu's, sure x86 platform today doesn't even have machines that have more than 64 cpu's outside of clusters, but in the next year or two that will change. While the base limitations of the x 86 platforms will remain, Intel and AMD are both working on multi-core CPU's, so once they get to 4 cores per CPU installed in a 32 CPU box, even x86 needs to scale to 128 cores. Is Linux up to the challenge? Perhaps there is room for another open source OS that can scale.
Another thing that Solaris has going for it is system and software monitoring, where most Linux sys-administrators use top or ps, too see what is happening with there systems, Solaris has top and ps available, but there are a lot more tools that are in the sys-admin arsenal as well. Sar, iostat, vmstat, prstat have a long history in Solaris, I know that all of those tools are ported to most open source OS's, but starting with Solaris 10, there is a new even more powerful tool to use, dtrace, is a scriptable monitoring tool, that can see even the lowest level of the kernel, it can do things that other OS's would require major changes of the kernel. One example of a script that is available for dtrace is called psio10, its like ps, but prints out IO usage on a task level so you can see how much data each application is generating; it does this while incurring very little use of the CPU. This is a small example; there are 100's others out there, and many more to come in the future.
Check out www.sun.com for more examples of features that may provide you with the tools to make you more successful.
Posted by James Dickens on December 22, 2004 at 04:24 AM JST #
Posted by Derek Morr on December 22, 2004 at 05:24 AM JST #
-dp, blogging this wirelessly from a Toshiba Tecra M1 running s10_71, using a NetGear MA-401 PCMCIA card.
Posted by Dan Price on December 22, 2004 at 05:57 AM JST #
If you know of instances where important customers have no idea what we're doing, please let us know (you could talk to Jim, or me). We can at least attempt to get someone dispatched to give them some useful information about what we've done, and where we're headed. If you have other ideas about publications we should target, etc., we'd like to hear it. Write a blog entry telling us what to do!
Thanks for being an advocate!
Posted by Dan Price on December 22, 2004 at 06:01 AM JST #
I think we should take stock of where we are right now (not get distracted by the past or the future), recognize where we are lacking or wrong and admit it (horror!), be proud of where we are strong, and stay positive as we move ahead aggressively. That's extremely difficult to do, I realize.
I love it when Solaris developers respond to my blog because it gives me more data points to understand the community we are trying to build. Sun needs this feedback (badly). And we *are* listening. Most of us, anyway. :) We need to remember that we do have a long way to go in certain markets, no question, but we are rock solid in other areas. I think the same could be said for Linux as well (thought in probably in different areas).
Linus' comments, however, point to a much larger issue that Jaime touches on to -- not enough people know we are here and what we have with Solaris 10, from a community perspective and from a product perspective. Although Sun is actively briefing developers, sys admins, customers, and partners regarding Solaris 10 and OpenSolaris, there's still a very long way to go. In this respect, I think Linus simply doesn't know what he's talking about. He's just not familiar with the technology we have in S10. Just as, in many instances, we didn't necessarily know what we were talking about with regard to Linux -- although we are learning more the last two years by engaging more with the community and integrating Linux systems into our product portfolio.
I'd love the competitive situation to be more focused on Unix vs Windows. What do you think?
Jim
Posted by Jim Grisanzio on December 22, 2004 at 07:32 AM JST #
Posted by Alan Hargreaves' Weblog on December 23, 2004 at 07:52 AM JST #
Could it be that Linus is simply worried about a repetition of the SCO lawsuits? SCO accused Linus and the whole Linux community of having "looked" to closely at the source code of another Unix.
And not just any other Unix, but the same Unix which Solaris 2 started with (Solaris was the reference implementation of SVR4). And didn't Sun recently pay SCO some big $$$ to "free" Solaris? And wasn't that money used in SCO's lawsuits?
That, and the fact that the CDDL is, by Sun's own words, not expected to be GPL compatible makes it doubtfull if looking at Solaris code is a good idea.
I at least would be worried to look at some code which has close ties to the tripple-S (SCO/Sun/Solaris). I would advice any Linux contributor to stay well away of OpenSolaris to avoid any potential conflict and to avoid a second SCO charade.
Maybe this is all paranoia. But I am not a Linux bigott (this posting is written on a noisy Blade 1500 :-), just someone who administered and worked on SunOS 4.x and Solaris for 15 years, and is sick and tired of Sun's talks and lack of action.
I think this is all about trust, or more precisely, the lack of trust in Sun and Solaris, not the quality or features of Solaris.
Posted by Erwin Lindner on January 18, 2005 at 06:07 PM JST #