Follow Slashdot blog updates by subscribing to our blog RSS feed

 



Forgot your password?
typodupeerror
×
The Almighty Buck Operating Systems BSD

HotJobs Upgrades to FreeBSD 49

bsdmike writes "DaemonNews has a link to an news article that reports that Yahoo! has saved something like $470,000 by switching HotJobs from Sun Solaris to FreeBSD. It's really amazing what affordable hardware and great Open Source technology can do!"
This discussion has been archived. No new comments can be posted.

HotJobs Upgrades to FreeBSD

Comments Filter:
  • by Kiwi ( 5214 ) on Sunday November 24, 2002 @03:57PM (#4745053) Homepage Journal
    Ironic that the trolls here post about how the BSDs are dying. Which they are not; all of the BSDs are being actively maintained and will continue to exist as long as the devlopers are dedicated to improving the BSD software.

    Sun, on the other hand, has the problem of needing to continue being a viable company. Otherwise, their OS will go the way of BeOS (impossible to maintian and update).

    In the last year, Sun has only posted a profit for one quarter. They lost $111 million last quarter alone and had to start another massive round of layoffs. Sun bet the bank on the dot-com boom continuing, and when that ballon popped, Sun found themsleves in a difficult position.

    The fact that all of the free OSes continue to improve in quality and scability does not bode well for Sun either. Oracle has recently replaced their very large Sun servers with arrays of Dell machines running a commodity OS.

    Sun has made a lot of contirbutions to free software, such as the RPC code, the NFS code, and Open Office. While not the most secure, these were valuable contributions which helped Linux and BSD become viable OSes in the early days of complete free software systems (early 1990s), and are making free *nix boxes more viable desktop machines today (with Open Office, though I use AbiWord myself).

    I hope Sun survives this attack; they will need to let go of their arrogance and their "The only real Unix is Solaris; Linux and *BSD are toys" mindset to survive.

    - Sam

    • Sun is on the wrong end of an economic trend. If they don't re-invent themselves, they'll become trapped in a slowly dieing (heh) nitch.

      The fact is, for the enterprise tasks most people would buy sun, these days the tasks parellelize well enough to run them on a cluster of commodity pc hardware. It isn't as reliable, but when you can afford a staggeringly larger number of resources, redundancy becomes very cheap.

      Much like vector supercomputers became impactical for most workloads, replaced by beowulf style clusters, sun' hardware will be relegated to areas where massively parellel shared memory architectures are key. The final nail in sun's coffin will be when an x86 chip can address large memories.
    • by Anonymous Coward on Sunday November 24, 2002 @04:30PM (#4745285)
      The fact that you think free UNIXes are comparable to Solaris is laughable and shows that you've never worked in a high end processing environment.
      • by Anonymous Coward
        My general observation is that Solaris advocates can not argue their way out of a paper bag. Usually, they respond to my comments about the shaky future of Solaris with personal insults and no facts whatsoever.

        The fact of the matter is that each and every version of Linux is becoming more and more like Solaris. Linux 2.6 (3.0?) will have even more stability, scability, an enterprise-ready LVM, and some other features such as kernel crash dumps (finally!, but only in unofficial patches). IBM, SGI, and other big players are investing serious money in to Linux; Solaris' days are numbered.

        Linux won for the same reason SunOS won over ten years ago. Comparing SunOS to offerings such as VMS 10 years ago was indeed "laughable". NFS was a joke compared to other network file systems at the time. RPC was also pathetic. Sun only won because they were the most open solution; the scability that Solaris advocates see Solaris as haivng today only came later.

        Sun is losing sales, and Linux is gaining sales, because Linux is more open than Solaris; not because it is better. And, just as sun did, Linux is becoming more and more scalable.

        - Sam
        • And this has always happened. The most open solution in the computer world comes along and beats the snot out of the competitors. DEC falls to Sun. Sun is falling to Microsoft in increasingly higher-end systems (to be fair, they don't aim for exactly the same markets), and Microsoft is falling to the Linux vendors and the BSD world.

          Sun may figure this out ahead of time -- they need to cut their profit margins, provide less expensive hardware, and avoid even the hint of trying to induce lock-in to their own services. However, I don't know if they can do it. Once you've been charging obscene amounts of money for a product for a long time, it's hard to stop and cut costs. People have to go, departments have to be streamlined...and there are hungry competitors that don't have all your baggage waiting...
          • Remember the browser war - cash rich company buying market share to create a monopoly, which they are successfully defending

            I think we (the open source zealots) tend to see everything the way we would like it to be, rather than is.

            The problem with the companies you list is, as you point out, that they have been attached to the teats of cash cows for too long and are vulnerable to disruptive innovations [hbs.edu]. You never hear the bullet that kills you.

            Part of the problem may be that previously small, nimble companies become bloated, their management structures become unable to prune the dying limbs, usually due to internal politics, and turf wars.

            It must be very difficult to fight a competitor (Linux) with no overheads, no fixed costs. Like a hydra - every time you attack it, it grows another distribution.

            It is strange that IBM doesn't figure in your list (no I don't work for them), as despite its monolithicity, it managed to see the Linux steamroller, and managed to get on board, rather than view the process from ground level. More credit to them, as it must have been a very courageous leap in the dark at the time.
      • Please tell, what is this mythical "high end processing environment" that I keep hearing about? Where exactly would a BSD-derived commercial Unix be better than a comparable BSD-derived free-software Unix? What can a Starfire server do that a cluster can't, and why is it better than a z-Series running a bunch of Linux images or some other mainframe?
        • Please tell, what is this mythical "high end processing environment" that I keep hearing about?

          I am not arguing with you. But the "high end processing environment" that I know Sun practically owns, is the financial sector. Stock Broking and Banking. Linux is coming up really fast on the desktop in the Stock Broking area though.

          Reality shows that technical details mean NOTHING in the corporate world. People in these sorts of environments continue to use what they trust, what they've trusted for many years. Products that have barely ever softened this trust.

          I have tried to deploy OpenBSD bridge firewalls and have been met with rejection by people with high technical prowess, because they fear that using OpenBSD for firewalls, moves the perceived accountability from, say Cisco, to themselves.

          Forget what the EULA's might say, forget that they (Cisco) most likely wash their hands of even the highest negligence and forget how technically brilliant the free solution might be.

          This situation of course, disgusts me.

          • I work in a trading firm, we're huge on Solaris. But we're also looking into Linux on IA32, IA64, and PowerPC (on a Power4 p690). Don't know what the migration will be like, but I can tell you a lot of folks are interested in the cost/speed advantages of Linux.

            One thing I think folks forget is that its not an all or nothing thing. Sure some folks need 9 nines, but a lot of folks that use Solaris RIGHT NOW don't. They can get away with Beowulf clusters and all that fun stuff. Can Sun live with this smaller customer base, those that need 9 nines and can't use the failovers and clusters available with Linux. Hmm, I don't know, interesting to see this play out.
        • Let's say: a telco's billing environment... Going for Sun gives you the option to buy 24/7 support on both hardware and software. Now I know some of you will argue that this is of limited or no use, or that the mailinglists and docs of BSD (or any other open product for that matter), still it's good to know that if something fails, I can get someone to replace it ASAP. You get the kind of assurance you want in this environment. Since you want the same assurances for you database, you end up with a vendor who will guarantee it will work. Limiting your options to a some Linux-es and commercial unixes... Let's take the Starfire example, it allows you to have a single database and make it do N million transactions/hour, something that will not be that easy on a cluster in the way it's needed (and trust me they've tried). Telcos have the tendency to want BIG suppliers for these things, just to make sure their supplier survives, and that they are solid enough to ensure they're around 3 years from now. Then there is the matter of culture, it's just now how it's done there... I'm not saying Solaris _will_ survive, I'm not saying BSD is not surviving, I'm not commenting on the technical differences... I'm not saying I think my observations are the way things should be, but it is how I see things are out there... I'm sorry, I wish things were better...
          • Absolutely.

            I've worked in quite a few Sun shops, and yes, being able to call Sun at 3 in the morning to replace that Enterprise box that just exploded on you YESTERDAY is a very nice thing to have. And not only Sun does that. SGI does that too. For a sysadmin, having working replacement hardware within a few hours is absolutely vital.

            This kind of service is hard to get from Free Software. Sure, the free Operating Systems are very good in quality, but the SLA's and support you get from the *nix vendors is just vital.

      • You are an idiot. The LLNL MCR system [llnl.gov] is running linux. Beats the heck out of most high end processing environments. Free UNIXes work great. There is sometimes a link between cost and ability - but not always.
    • Oracle has recently replaced their very large Sun servers with arrays of Dell machines running a commodity OS
      Yes, Solaris likely is in more danger than the BSDs; if Sun keeps decreasing in viability. Solaris may go away. However, IMO Sun's problem is on the hardware side. Sun makes great hardware, and Solaris on Sun hardware is a tough combination to beat. But Sun hardware is expensive, and cheap x86 systems can run circles around Sun hardware at the same price point, so this is where Sun is in danger. If SPARC hardware ever goes away, Solaris could likely go away with it.

      ...their "The only real Unix is Solaris; Linux and *BSD are toys" mindset...
      All I can say is that on SPARC hardware, Solaris really is the best OS. OpenBSD and NetBSD run nicely on SPARC (it's a very mature port,) but IME Linux is sketchy at best on SPARC. In the past Sun has criticized Linux (and rightfully so as it applies to SPARC hardware) but I don't recall any negativity towards OpenBSD or NetBSD.
    • Sun is a HARDWARE company. Their hardware ROCKS. Soon, FreeBSD will run on Sun hardware. When it does, Solaris will not be threatened. Big database vendors will continue to support Solaris because it is the pet OS of the very smart Sun engineers who produce the *Sparc hardware platforms. I use FreeBSD at home, and wherever I can at work, but I use Solaris at work to provide the juice for our big databases. It goes real fast. It goes on boxen with 4-64 processors. FreeBSD has a while to catch up with that.

      I'm not sure if this guy is a troll and I should have refrained from this post, but hopefully people will get something out of my response. Turning lemons into lemonade y'know..

How many QA engineers does it take to screw in a lightbulb? 3: 1 to screw it in and 2 to say "I told you so" when it doesn't work.

Working...