Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 



Forgot your password?
typodupeerror
×
Sun Microsystems Operating Systems Software IT

Sun to Make Solaris More Linux Like 400

ramboando writes "In an effort to spur adoption of Solaris, Sun Microsystems has begun a project code-named Indiana to try to give its operating system some of Linux's success. Sun has been trying for years to restore the luster of Solaris, but that since has faced a strong challenge chiefly from Linux. Sun wants to embrace some Linux elements so "we make Solaris a better Linux than Linux," said Ian Murdock, Sun's chief operating systems officer, quoting Netscape co-founder Marc Andreessen, whose latest start-up, Ning, uses Solaris. But it's a tricky balance to adopt elements of Linux while preserving Solaris technology and advantages such as the promise of backward compatibility. "As we make Solaris more familiar to Linux users, we don't [want to] lose what makes it more compelling and competitive.""
This discussion has been archived. No new comments can be posted.

Sun to Make Solaris More Linux Like

Comments Filter:
  • by AKAImBatman ( 238306 ) * <[moc.liamg] [ta] [namtabmiaka]> on Thursday May 10, 2007 @12:47AM (#19062935) Homepage Journal
    Not to say that some of the Solaris tools couldn't use a good sprucing up with newer and fresher versions, but I tend to get nervous whenever Sun codenames something. It usually means that they're about to start on something that isn't a bad idea per se, but will be guaranteed to be aborted prior to any real commitment or follow-through. What state that will leave Solaris in is anyone's guess.

    *shudder* I still remember Mad Hatter. Such promise. Such failure to follow up,
  • by koreth ( 409849 ) * on Thursday May 10, 2007 @12:48AM (#19062947)
    I've liked many aspects of Solaris for a long time, but the #1 thing that turns me off it is the userland tools.

    Yes, I know they ship a DVD with lots of GNU tools, but the fact that the built-in make, vi, grep, etc. are still basically unmodified from the early 1990s (if not longer) is not, to me, a feature. Those hoary old versions should be the ones on a supplementary DVD for those who need perfect backward compatibility with 15-year-old shell scripts and so forth.

    It sounds like that's a focus of this project, so I say fabulous. If I can get ZFS and DTrace plus a modern toolset out of the box, Solaris will start to look much more attractive.
  • Err.... (Score:2, Insightful)

    by Mr. Flibble ( 12943 ) on Thursday May 10, 2007 @12:50AM (#19062967) Homepage
    Or you could just run Linux on Sun hardware?

    Sun is hemmoraging cash. Their hardware is fairly standard (in an enterprise way) and all the functionality of Linux has jumped ahead of Solaris... So what do they have to offer? Nothing. I can't see what they can do in this regard to gain back market share. making a "better linux" than Linux is not it.

    There are probably other paths that they can take that would be more effective than this one. But I don't know what they are.
  • Business model? (Score:5, Insightful)

    by Urusai ( 865560 ) on Thursday May 10, 2007 @01:00AM (#19063029)
    Sun is a for-profit entity. How do they expect to make money off of their OS? They should GPL Solaris, let the code monkeys snatch the best bits for Linux, and forget about wasting their money developing Solaris. They can write a "shim layer" on Linux for people needing backward compatibility so they don't alienate long-time customers. They need to figure out where they plan on making money, and scrap the parts that lose money. Open sourcing Java was an indication of desperation; we saw plenty of companies open source their product during the dot-com bust, either because they didn't want their work to die, or because they thought it would magically boost market share and generate revenue. It doesn't.
  • by misleb ( 129952 ) on Thursday May 10, 2007 @01:02AM (#19063043)
    Yeah, and OS/2 was a better Windows than Windows. Anyone remember how that worked out?

    -matthew
  • by Anonymous Coward on Thursday May 10, 2007 @01:15AM (#19063125)
    Yeah, that project codenamed Oak was pretty much a bust.
     
    /sarcasm
  • by NeuroManson ( 214835 ) on Thursday May 10, 2007 @01:19AM (#19063147) Homepage
    But what exactly makes Solaris worth using to begin with? What open source or commercial software makes it worth having? What makes it more than just a fringe system? Linux is finally approaching the point where it stands a chance at competing against Windows in the consumer market, does it really need competition from a fairly mainstream corporation?

    For that matter, sure, the machines look cool on the outside, but why do so many people consider them worth buying (even models up to 10 years old) today, and for that matter, what makes them worth switching over to? Is it sheer geek chic, or do they actually provide some form of useful function, as opposed to Windows/Mac/Linux's growing trend towards multipurpose multimedia machines?
  • by JonathanR ( 852748 ) on Thursday May 10, 2007 @01:20AM (#19063163)

    Sun has been trying for years to restore the luster of Solaris, but that since has faced a strong challenge chiefly from Linux

    "As we make Solaris more familiar to Linux users, we don't [want to] lose what makes it more compelling and competitive."
    If it is "more compelling and competitive" [than some other OS, whichever that is], then why the obsession with following after Linux? If Solaris is on the decline, then why not suspend further Solaris development, and launch their own Linux distro along side?
  • by advocate_one ( 662832 ) on Thursday May 10, 2007 @01:31AM (#19063247)
    that's all we want... the current list of supported x86 hardware is ridiculously small... oh and put some effort into Gnu/Solaris... that project has effectively stagnated for ages now and nothing appears to be happening...
  • by Coryoth ( 254751 ) on Thursday May 10, 2007 @01:38AM (#19063275) Homepage Journal

    "but the fact that the built-in make, vi, grep, etc. are still basically unmodified"

    Who cares? Do they work?
    That depends on your measure of "work". They do the raw bare minimum one would expect from such things, but the GNU versions tend to come with a lot of comforts that you start taking for granted after not very long. Its nothing you can't technically live without, but it does start to feel awfully spartan. A good comparison might be Solaris grep [sun.com] and GNU grep [ed.ac.uk], or perhaps Solaris diff [sun.com] and GNU diff [ed.ac.uk]. Nothing wrong with the Solaris versions, but the GNU versions have some useful extra options, and more flexible regexps.
  • by giarcgood ( 857371 ) on Thursday May 10, 2007 @01:42AM (#19063297)

    oh and put some effort into Gnu/Solaris... that project has effectively stagnated for ages now and nothing appears to be happening...


    Yes Sir! Anything else you would like for free?
  • Re:Err.... (Score:5, Insightful)

    by Coryoth ( 254751 ) on Thursday May 10, 2007 @01:50AM (#19063335) Homepage Journal

    I thought you could already use DTrace on Linux, and if they GPL their stuff, it will all be ported to Linux. The article says that it would be hard, but you know it would happen.
    Linux does not have DTrace. You're right that it will probably happen. Eventually, after much work. And ZFS isn't looking like it'll be an easy addition either. There doesn't look to be an equivalent of Zones either -- Linux has some nice security module hook in the kernel thanks to work by the NSA, but right now it is largely unused (even distros that enable SELinux have very lax policies, and fairly basic management). Again, that might arrive, at some indeterminate time in the future. Considering that your original post was proclaiming:

    ...and all the functionality of Linux has jumped ahead of Solaris...
    arguing that Linux may eventually catch up with these powerful Solaris features is a little disingenuous don't you think? Linux and Solaris are both worth having, depending on what you need. I look forward to what this project, and the OpenSolaris project, can put together.
  • Comment removed (Score:3, Insightful)

    by account_deleted ( 4530225 ) on Thursday May 10, 2007 @01:54AM (#19063357)
    Comment removed based on user account deletion
  • by Anonymous Coward on Thursday May 10, 2007 @02:06AM (#19063417)
    No baby, it's just because of JavaOne 2007 - Java Developers Conference which makes successive new product and technology announcements.
  • by timmarhy ( 659436 ) on Thursday May 10, 2007 @03:16AM (#19063811)
    solaris is REAL enterprise sector stuff. they don't give a shit about lastest and greatest, they care about stability and basic functionality.

    try running ubuntu on a fortune 500 companys network and see how you fair.

  • by Asmodai ( 13932 ) on Thursday May 10, 2007 @03:32AM (#19063885) Homepage
    Oh please. Lay off the bash fanboyism already. I personally get sick and tired of scripts that assume bash to have been installed under /bin. At least use a more portable hash-bang sequence like #!/usr/bin/env bash to make them semi-portable. Make the default shell a normal bourne again shell and allow users to switch to their own preferred one.

    Also if the bash manual page says this:

    BUGS
                  It's too big and too slow.

    Then you just know it is a bad choice beyond even other considerations.
  • by Skrynesaver ( 994435 ) on Thursday May 10, 2007 @03:40AM (#19063915) Homepage
    You should look at you kernel parameters ulimit -a As shipped Solaris is intended for big iron in a way that most Linux distros aren't
  • by Anonymous Coward on Thursday May 10, 2007 @05:20AM (#19064435)

    We named the dog 'Indiana'.

  • by Jon Peterson ( 1443 ) <jon@@@snowdrift...org> on Thursday May 10, 2007 @06:23AM (#19064685) Homepage
    Although I'm not hands on now, I originally moved from Linux to Solaris (with some Irix stuff in the middle). I still prefer Solaris for the following reason:

    Simple is better.

    This single thought is perhaps the biggest lesson I've learned in my whole career, about almost any aspect of computing. Complexity is the enemy.

    caveat: by 'Linux' I mean 'The particular distro your company has standardised on'
    caveat: I'm only concerned with servers. Solaris may be the worst desktop OS in the world FAIK.

    1. Less shovelware. Although a base Solaris install is still annoyingly large, it's not nearly as bad as most Linux distros. It infuriates me that operating systems think its useful to install entire database, programming languages, you name its 'just in case you need them'.
    2. Better backward compatability. Upgrades to discreet parts of Solaris don't usually require upgrades to other parts of Solaris. This means that you aren't constantly trying to run the latest versions of everything.
    3. Better hardware integration. When you are running a lot of servers, it's very useful to have a nice console, so you can talk to the things properly. I think Linux has improved a bit in this area, but I'm not aware that it has an equivalent to the OK prompt, and the various diagnostic tools therein.

    Others have talked about various tools and kernel level stuff, but I wanted to make that point that while the Solaris userland might feel archaic to some, to me it feels pleasantly simple - devoid of hidden complexity, obscure features that badly written apps come to rely on, and all the other 'let's have another feature' attitude prevalent in much OS software.

    To me, Solaris feels like HTTP, and Linux feels like SOAP.
  • by Anonymous Coward on Thursday May 10, 2007 @06:30AM (#19064731)
    Anonymous Coward for a reason: I'm a Sun TSE.


    Caseih" is correct when he says "This certainly isn't quite the same Sun as in the olden days", in regard to support and how it is delivered. It certainly isn't the same Sun for those of us who are tasked with delivering support. Management has implemented all sorts of programs to improve customer "sat" and bring down call hold times, programs that INTERFERE with the day to day support work; effective and seasoned TSEs are bailing out right and left and ARE NOT BEING REPLACED in many cases; the EDS "partners" have a large turnover rate (what do you want for $9 an hour?); more time on the phone taking live calls, meaning the TSE have less (or no) time to do followups, research, spend time in the lab . . .. . . . I could go on but you get the idea.


    The "Dell-ization" of tech support is spreading like a virus; support is a commodity now. Even enterprise level tech support. Sold to the lowest bidder. Who cares if the person on the phone can't spell "LDAP", as long as the call is picked up in X minutes and keeps the manager's pager from going off? THAT is where Sun support is today.

  • Re:Err.... (Score:3, Insightful)

    by Rich0 ( 548339 ) on Thursday May 10, 2007 @06:47AM (#19064813) Homepage
    The whole point of the no-binary-interface to the kernel is to make your life as big of a pain as possible. That discourages you from buying stuff from nvidia/etc. Or at least making a lot of noise. That encourages other vendors to open-source their drivers and integrate them into the kernel.

    If anybody could easily maintain binary kernel drivers then more people would do so, and less code would be released open-source. Now, the total amount of code (proprietary+otherwise) might increase, but that isn't really Linus's concern. If the code isn't open it doesn't really benefit him. When you think about it, more people using linux doesn't really benefit him either...
  • by swordgeek ( 112599 ) on Thursday May 10, 2007 @07:04AM (#19064903) Journal
    I'm a full time Solaris admin, and getting paid well for it. If supply and demand changes to get idiots like you out of the field, then I'd be happy.

    Good, professional admins will always command a premium. There are very VERY few good, professional Linux or Windows admins, but they're paid roughly as well as Solaris (or HP-UX, etc.) admins.

    "any monkey thats run linux for a bit has the same skills?"

    I suggest you upgrade your skills beyond those of a monkey. And your attitude, while you're at it.
  • by Junta ( 36770 ) on Thursday May 10, 2007 @08:17AM (#19065395)
    Well, to one part of your argument, I know at least one fortune 100 company that has a fair amount of ubuntu in place. I also know that a fortune 1 company uses linux extensively, though I do not have specifics on what architecture. With HP, Sun, IBM, et. al taking linux Damn seriously, you can consider linux *real enterprrise sector stuff*. Linux is popular because it implements the fundamental design of Unix systems in a development situation that largely precludes any sort of vendor lock-in. You buy AIX on IBM System p, you have committed that as time goes by your investment is tied to buying from IBM again. You buy RedHat on Dell, and Dell disappoints you can go try HP in a future upgrade with minimal changes (one x86 box is just like another for most fundamental ways that matter). If Red Hat pisses you off, you go to Novell (not quite as non-impact, but certainly well within the realm of possibility, better than, say, AIX to HP-UX). Technical people love the unix-like architecture and the ready availability for whatever they wish. Business loves linux because of the vendor freedom and because the technical guys who love it and know it well are plentiful. Any interview I conduct, I ask about home usage and what they are looking into outside the boundaries of commercial experience. Inevitably the answers are more technically advanced and prove qualifications beyond their commercial work. Being freely available has not hurt. Solaris absolutely will need to cede control and authority so that more than one healthy commercial vendor sells and can support Solaris 100% independent of Sun's help. Making it supported on non Sun systems and x86 didn't help, making it as free-as-in-beer for most people didn't help, and making it more BSD-like has yet to make significant progress. If they GPL the codebase I don't think that in and of itself will help, but if some company or two succeeds in becoming a prominent solaris vendor who doesn't have to go to Sun for any partnership or anything, then it could begin to work, but they still have the momentum of linux which is not a situation easily overcome. I do think if they succeeded in making Solaris a prominent platform, their commercial distribution of it would probably not be that popular (I don't think on many fronts Sun 'gets it' on some of the technical things not right out-of-the-box with their software, the core is good and a good system can be built on it, but I don't think Sun is capable). Admittedly a small market share of a linux-scale market is much better than their total market-share of a small market.

    Now, even if your statement was 100% accurate in every sense of the word, Nexenta's lack of development does *not* represent a stable and basically functional system. It represents a stale Nevada build. Sun has done many better builds since the last Nexenta release. A pity, Nexenta debian-ified Solaris enough to have the package management and general interface strategy be bearable (No matter how you slice it, Nevada's UI may have better options, but it's still ugly and misses a lot of the point in my opinion.
  • by ajs318 ( 655362 ) <(ku.oc.dohshtrae) (ta) (2pser_ds)> on Thursday May 10, 2007 @08:46AM (#19065673)
    And so, the wheel turns full circle. As summer fades into autumn, then winter gives way to spring and summer returns again; so doth GNU depart from Unix, only to return again to Unix.

    The GNU project was originally meant to be an alternative to the closed-source Unix implementations of the day. Like a heroin dealer relying on the twin pillars of illegality and addictive potential, closed-source Unix vendors had little incentive to improve their products; they just had to be different enough from the competition that you couldn't switch easily.

    It really took for Linux to come on the scene to get GNU into a usable state; the BSD kernel (which had been favoured by the GNU developers prior to the advent Linux) already came with well-matched userland tools. And you've got to be serious about something to buy a whole car that already works just to rip out the engine and use it in a different chassis that looks identical to the first one from a distance. The GNU/Linux combination sparked interest in GNU. In turn, the BSDs diversified; today FreeBSD, OpenBSD and NetBSD all have their own respective market niches.

    Closed-source Unix continues to stagnate and ultimately will grow irrelevant. The elephant in the room is that neither hardware nor software make up the bulk of the intrinsic value of a computer system; that value comes mainly from users' saved data.

    Open Source pretty much forces you to implement Open Standards for saved files, which leads to transparent interoperability between programs that do the same sort of thing. In the end, AbiWord on GNU/Linux, OpenOffice.org on Solaris and KWord on FreeBSD will all be able to open the same documents. The brand of tools used to shape the data is becoming less important than the result of using them. That's already how it is in other industries. After all, who ever asked what brand of cooking equipment a restaurant uses, or what make of tools a cabinet maker uses? The important thing is that chopping food with one make of knife doesn't block you from cooking it in a different manufacturer's pans, and rough-cutting a piece of wood with one make of power saw doesn't prevent you finishing it with a different manufacturer's chisel. Using one OS and application stack on your computer shouldn't preclude you from working with data manipulated using a different OS and stack. That's already the way it's heading, slowly but surely.
  • Nextenta! (Score:3, Insightful)

    by MoxFulder ( 159829 ) on Thursday May 10, 2007 @09:23AM (#19066105) Homepage
    That's pretty neat, thanks! I'd never heard of Nexenta before. So basically it's the kernel+libc from Solaris, with the Debian userland...

    But, uhm, is there any real evidence that the Solaris kernel is actually *better* than the Linux kernel? The Linux kernel definitely supports a LOT more hardware. Although Solaris is seen as more heavy duty by a lot of IT folks, I'm not sure if there's a good reason for this besides long-time familiarity.
  • If you're as hardcore as I am, you realize that Sun OS hasn't been a true operating system since the day they unbundled the compiler. An OS without a compiler? What do they expect me to use? Stone Knives and bear skins?
  • by 5pp000 ( 873881 ) on Thursday May 10, 2007 @10:52AM (#19067581)

    +5, Insightful? Oh, come on, mods.

    This is not a problem that can be tuned away. I'll tell you exactly what's going on. The kernel has, for each process, a table which contains one entry for each contiguous region of address space with the same page protections. Since the Lisp implementation I'm using makes use of page protections to implement its GC write barrier -- a very useful technique for an SMP garbage collector -- it creates lots of small regions, so that this table gets quite large. And, there are algorithms in the kernel that are quadratic, or worse, in the size of the table.

    The result is that as the Lisp heap grows past a couple of GB, one of the CPUs (I'm doing this, BTW, on a quad Opteron with 16GB of DRAM) comes to spend 100% of its time in the kernel, doing whatever this quadratic algorithm is doing, and the machine becomes pretty much unresponsive.

    Solaris has no trace of this behavior. Clearly, it comes from a culture where the OS is expected to scale in many dimensions, and quadratic algorithms are strictly forbidden.

"I don't believe in sweeping social change being manifested by one person, unless he has an atomic weapon." -- Howard Chaykin

Working...